[ { "moduleName": "default", "type": "image", "isEssential": true, "category": "生成类型", "icon": "icon-[lucide--star]", "title": "使用案例", "templates": [ { "name": "template_seedance_2_0_plus_llm_prompt_helper", "title": "Seedance 2.0 + LLM 提示助手", "description": "上传两张参考图片并输入简短文字描述。使用LLM助手自动生成详细的Seedance 2.0风格提示。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "视频参考" ], "models": [ "Gemini-3.1-Pro", "Seedance 2.0" ], "logos": [ { "provider": [ "Google", "ByteDance" ] } ], "date": "2026-05-10", "openSource": false, "size": 0, "vram": 0, "usage": 420, "searchRank": 0, "username": "enigmatic_e", "io": { "inputs": [ { "nodeId": 352, "nodeType": "LoadImage", "file": "freezer.png", "mediaType": "image" }, { "nodeId": 356, "nodeType": "LoadImage", "file": "snowboarder.png", "mediaType": "image" } ] } }, { "name": "api_seedance2_0_r2v", "title": "Seedance 2.0：视频参考生成", "description": "根据参考图像和文本提示生成电影级视频。在保留主体身份和构图的同时，通过同步音频添加富有表现力的运动。通过详细提示控制摄像机移动和灯光。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "图生视频", "视频" ], "models": [ "Seedance 2.0" ], "logos": [ { "provider": "ByteDance" } ], "date": "2026-05-25", "openSource": false, "size": 0, "vram": 0, "usage": 1753, "searchRank": 1000, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 352, "nodeType": "LoadImage", "file": "sd_outfit_1.png", "mediaType": "image" }, { "nodeId": 353, "nodeType": "LoadImage", "file": "sd_outfit_2.png", "mediaType": "image" }, { "nodeId": 354, "nodeType": "LoadImage", "file": "sd_outfit_3.png", "mediaType": "image" }, { "nodeId": 356, "nodeType": "LoadImage", "file": "selfie_cartoon_style.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 344, "nodeType": "SaveVideo", "file": "Seedance2.0_r2v_00006_.mp4", "mediaType": "video" } ] }, "thumbnail": [ "output/Seedance2.0_r2v_00006_.mp4" ] }, { "name": "template_seedance2_0_viral_videos_character_swap", "title": "Seedance 2.0: Viral Videos Character Swap", "description": "A one click workflow to recreate an input video with your character. All you need to do is upload the video and your character. This workflow uses Google Gemini LLM to analyze the input video and reverse engineer a prompt. You can prompt for specific details to inject into this analysis. The workflow extracts the first frame of the video and swaps your character into it, sends this into Seedance 2.0 Reference-to-Video and generates", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "hoverDissolve", "tags": [ "API", "换脸", "图生视频" ], "models": [ "GPT Image 2", "Google Gemini", "Seedance 2.0" ], "logos": [ { "provider": [ "Google", "OpenAI", "ByteDance" ], "label": "Multiple providers" } ], "date": "2026-05-19", "openSource": false, "size": 0, "vram": 0, "usage": 201, "searchRank": 0, "username": "hellorob", "io": { "inputs": [ { "nodeId": 18, "nodeType": "LoadVideo", "file": "source_ref_video.mp4", "mediaType": "video" }, { "nodeId": 29, "nodeType": "LoadImage", "file": "virtual_character_ref.png", "mediaType": "image" } ] } }, { "name": "api_seedance2_0_t2v", "title": "Seedance 2.0：文生视频", "description": "通过文本提示生成电影级视频，具备同步音频、可控相机运动和稳定视觉效果。输入文本描述即可输出高质量视频序列。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生视频", "视频" ], "models": [ "Seedance 2.0" ], "logos": [ { "provider": "ByteDance" } ], "date": "2026-05-02", "openSource": false, "size": 0, "vram": 0, "usage": 288, "searchRank": 1000, "username": "ComfyUI", "io": { "outputs": [ { "nodeId": 344, "nodeType": "SaveVideo", "file": "api_seedance2_0_t2v.mp4", "mediaType": "video" } ] }, "thumbnail": [ "output/api_seedance2_0_t2v.mp4" ] }, { "name": "api_seedance2_0_flf2v", "title": "Seedance2.0：首尾帧视频", "description": "上传视频的首尾帧。使用Seedance2.0模型生成高质量视频序列。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "视频", "首尾帧视频", "API" ], "models": [ "Seedance 2.0" ], "logos": [ { "provider": "ByteDance" } ], "date": "2026-04-13", "openSource": false, "size": 0, "vram": 0, "usage": 731, "searchRank": 500, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 3, "nodeType": "LoadImage", "file": "sd_first_frame.png", "mediaType": "image" }, { "nodeId": 4, "nodeType": "LoadImage", "file": "sd_last_frame.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 2, "nodeType": "SaveVideo", "file": "api_seedance2_0_flf2v.mp4", "mediaType": "video" } ] }, "thumbnail": [ "output/api_seedance2_0_flf2v.mp4" ] }, { "name": "api_seedance2_0_flf2v_real_human", "title": "Seedance2.0 真人：FLF2V", "description": "上传一张真人参考图像进行身份验证，并使用Seedance2.0 FLF2V生成具有首帧和尾帧控制的个性化视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "视频" ], "models": [ "Seedance 2.0" ], "date": "2026-04-23", "openSource": false, "size": 0, "vram": 0, "usage": 38, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 20, "nodeType": "LoadImage", "file": "flf2v_demo_first_frame.png", "mediaType": "image" }, { "nodeId": 38, "nodeType": "LoadImage", "file": "flf2v_demo_last_frame.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 15, "nodeType": "SaveVideo", "file": "api_seedance2_0_flf2v_real_human.mp4", "mediaType": "video" } ] }, "thumbnail": [ "output/api_seedance2_0_flf2v_real_human.mp4" ], "logos": [ { "provider": "ByteDance" } ] }, { "name": "api_seedance2_0_r2v_real_human", "title": "Seedance2.0 真人：参考视频", "description": "上传真人参考视频进行身份验证。使用自己的照片验证并生成个性化视频输出。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "视频" ], "models": [ "Seedance 2.0" ], "logos": [ { "provider": "ByteDance" } ], "date": "2026-04-23", "openSource": false, "size": 0, "vram": 0, "usage": 123, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 1, "nodeType": "LoadImage", "file": "real_human_demo.png", "mediaType": "image" } ] }, "thumbnail": [ "output/real_human_demo.mp4" ] }, { "name": "templates-1_click_multiple_character_angles-v1.0", "title": "一键多角度角色", "mediaType": "image", "mediaSubtype": "webp", "description": "上传角色图片，获取多角度视图", "tags": [ "图片编辑" ], "models": [ "Qwen-Image-Edit" ], "date": "2025-12-08", "searchRank": 8, "size": 31245887078, "vram": 31245887078, "usage": 497, "io": { "inputs": [ { "nodeId": 25, "nodeType": "LoadImage", "file": "character_input.png", "mediaType": "image" } ] }, "username": "hellorob", "thumbnail": [ "thumbnail/templates-1_click_multiple_character_angles-v1_0.png" ], "openSource": true }, { "name": "template_qwen_image_edit_2511_systms_action", "title": "Qwen Image Edit 2511: SYSTMS ACTION LoRA", "description": "Upload your target image and use the trigger phrase \"action the [thing]\" (e.g., \"action the scene\"). Generate a toy or action figure version of your subject with articulated joints, plastic materials, and painted details.", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "图片编辑", "LoRA" ], "models": [ "Qwen Image Edit 2511" ], "date": "2026-05-10", "openSource": true, "size": 51646981734, "vram": 51646981734, "usage": 0, "searchRank": 0, "username": "ingi", "io": { "inputs": [ { "nodeId": 223, "nodeType": "LoadImage", "file": "western_rider.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 9, "nodeType": "SaveImage", "file": "template_qwen_image_edit_2511_systms_action.png", "mediaType": "image" } ] }, "thumbnail": [ "input/western_rider.png", "output/template_qwen_image_edit_2511_systms_action.png" ], "includeOnDistributions": [ "local" ] }, { "name": "template_ltx2_3_style_transition", "title": "LTX-2.3 风格过渡", "description": "应用ltx2.3-transition LoRA以创建流畅的风格和场景过渡。此工作流程使用专用模型来融合视觉元素。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "首尾帧视频", "LoRA", "视频" ], "models": [ "LTX-2.3" ], "date": "2026-03-30", "openSource": true, "size": 39406324941, "vram": 39406324941, "usage": 63, "searchRank": 0, "username": "shane", "io": { "inputs": [ { "nodeId": 137, "nodeType": "LoadImage", "file": "ltx23_flf2v_last_frame.png", "mediaType": "image" }, { "nodeId": 138, "nodeType": "LoadImage", "file": "ltx23_flf2v_first_frame.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 68, "nodeType": "SaveVideo", "file": "template_ltx2_3_style_transition.mp4", "mediaType": "video" } ] }, "thumbnail": [ "thumbnail/template_ltx2_3_style_transition.mp4" ] }, { "name": "template_qwen_Image_2512_360_lora", "title": "Qwen Image 2512：360度全景图像", "description": "使用排名128的LoRA在20B MMDiT模型上，从文本描述生成360度等距柱状投影图像。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "LoRA", "文生图" ], "models": [ "Qwen-Image 2512" ], "date": "2026-05-19", "openSource": true, "size": 32534377267, "vram": 32534377267, "usage": 99, "searchRank": 0, "username": "ComfyUI" }, { "name": "template_contact_sheet-step_1.app", "title": "步骤 1：联系表工作流程", "description": "上传您的角色、场景和产品。将所有图像整合成时尚大片风格的照片。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图像", "图片编辑" ], "models": [ "Nano Banana Pro" ], "logos": [ { "provider": "Google" } ], "date": "2026-03-30", "openSource": false, "size": 0, "vram": 0, "usage": 30, "searchRank": 0, "username": "hellorob", "io": { "inputs": [ { "nodeId": 1, "nodeType": "LoadImage", "file": "snake_on_jacket.jpg", "mediaType": "image" }, { "nodeId": 2, "nodeType": "LoadImage", "file": "forest_background.jpg", "mediaType": "image" }, { "nodeId": 5, "nodeType": "LoadImage", "file": "brown_moisturizer_bottle.jpg", "mediaType": "image" } ] }, "thumbnail": [ "thumbnail/template_contact_sheet-step_1.png" ] }, { "name": "template_contact_sheet-step_3.app", "title": "步骤 3：联系表工作流程", "description": "上传8张图片（按顺序）作为7个Kling 3.0 FL2V视频的关键帧。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "首尾帧视频", "视频" ], "models": [ "Gemini-3.1-flash-lite", "Kling 3.0" ], "logos": [ { "provider": [ "Google", "Kling" ] } ], "date": "2026-03-30", "openSource": false, "size": 0, "vram": 0, "usage": 18, "searchRank": 0, "username": "hellorob", "io": { "inputs": [ { "nodeId": 224, "nodeType": "LoadImage", "file": "contact_sheet_app_input_1.png", "mediaType": "image" }, { "nodeId": 226, "nodeType": "LoadImage", "file": "contact_sheet_app_input_2.png", "mediaType": "image" }, { "nodeId": 227, "nodeType": "LoadImage", "file": "contact_sheet_app_input_3.png", "mediaType": "image" }, { "nodeId": 228, "nodeType": "LoadImage", "file": "contact_sheet_app_input_4.png", "mediaType": "image" }, { "nodeId": 232, "nodeType": "LoadImage", "file": "contact_sheet_app_input_5.png", "mediaType": "image" }, { "nodeId": 233, "nodeType": "LoadImage", "file": "contact_sheet_app_input_6.png", "mediaType": "image" }, { "nodeId": 234, "nodeType": "LoadImage", "file": "contact_sheet_app_input_7.png", "mediaType": "image" } ] }, "thumbnail": [ "thumbnail/template_contact_sheet-step_3.mp4" ] }, { "name": "template_character_portrait_relighting", "title": "人物肖像重新打光", "description": "上传角色肖像，在保持角色一致性的同时创建多种电影级光照条件。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "重打光", "肖像", "API" ], "models": [ "Nano Banana Pro" ], "logos": [ { "provider": "Google" } ], "date": "2026-04-06", "openSource": false, "size": 0, "vram": 0, "usage": 84, "searchRank": 0, "username": "hellorob", "io": { "inputs": [ { "nodeId": 124, "nodeType": "LoadImage", "file": "silver_hand.png", "mediaType": "image" } ] } }, { "name": "templates_graphic_design_recomposer", "title": "图形设计重组器", "description": "上传图形设计并选择所需宽高比。生成并重新构图以适应新的宽高比。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "布局设计" ], "models": [ "Nano Banana 2", "Nano Banana Pro" ], "logos": [ { "provider": "Google" } ], "date": "2026-03-30", "openSource": false, "size": 0, "vram": 0, "usage": 25, "searchRank": 0, "username": "hellorob", "io": { "inputs": [ { "nodeId": 125, "nodeType": "LoadImage", "file": "latent_space_clud.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 131, "nodeType": "SaveImage", "file": "templates_graphic_design_recomposer.png", "mediaType": "image" } ] }, "thumbnail": [ "thumbnail/templates_graphic_design_recomposer.mp4" ] }, { "name": "template_image_speech_to_video", "title": "使用语音克隆生成UGC视频", "description": "上传图像以生成AI提示并合成语音。使用Eleven Labs和LTX Video创建具有准确口型同步的最终视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "口型同步", "视频生成" ], "models": [ "ElevenLabs", "LTX-2.3" ], "logos": [ { "provider": [ "ElevenLabs", "Google", "Lightricks" ], "label": "Multiple Providers" } ], "date": "2026-04-06", "openSource": false, "size": 48962627174, "vram": 48962627174, "usage": 49, "searchRank": 0, "username": "sirolim", "io": { "inputs": [ { "nodeId": 440, "nodeType": "LoadImage", "file": "template_image_speech_to_video_woman_holding_face_oil.png", "mediaType": "image" }, { "nodeId": 611, "nodeType": "LoadAudio", "file": "voice_demo.mp3", "mediaType": "audio" } ], "outputs": [ { "nodeId": 479, "nodeType": "SaveVideo", "file": "template_image_speech_to_video.mp4", "mediaType": "video" } ] } }, { "name": "template_product_placement", "title": "产品展示", "description": "将产品替换到参考场景中，同时保持构图和光照。生成最终产品图像和自动视频提示，用于创建精美的产品视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "产品", "替换", "视频" ], "models": [ "Gemini-3.1-Pro", "Kling 3.0", "Nano Banana 2" ], "logos": [ { "provider": [ "Google", "Kling" ] } ], "date": "2026-03-30", "openSource": false, "size": 0, "vram": 0, "usage": 34, "searchRank": 0, "username": "Julien | MJM", "io": { "inputs": [ { "nodeId": 1, "nodeType": "LoadImage", "file": "product_1_product_placement.png", "mediaType": "image" }, { "nodeId": 2, "nodeType": "LoadImage", "file": "product_2_product_placement.png", "mediaType": "image" }, { "nodeId": 3, "nodeType": "LoadImage", "file": "product_3_product_placement.png", "mediaType": "image" }, { "nodeId": 5, "nodeType": "LoadImage", "file": "reference_image_product_placement.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 10, "nodeType": "SaveVideo", "file": "template_product_placement.mp4", "mediaType": "video" } ] } }, { "name": "template_qwen_image_illustration_lora", "title": "Qwen Image：插画LoRA", "description": "基于Qwen-Image构建的风格LoRA的t2v工作流。生成涵盖可爱动漫角色设计、欧洲连环漫画、独立孔版印刷、故事书水彩、复古少女漫画和图像小说插图的图像。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "LoRA", "文生图" ], "models": [ "Qwen-Image" ], "date": "2026-03-30", "openSource": true, "size": 30816390349, "vram": 30816390349, "usage": 95, "searchRank": 0, "username": "ComfyUI", "io": { "outputs": [ { "nodeId": 60, "nodeType": "SaveImage", "file": "template_qwen_image_illustration_lora.png", "mediaType": "image" } ] }, "thumbnail": [ "output/template_qwen_image_illustration_lora.png" ] }, { "name": "template_sugar_coated_gummy_style_qwen", "title": "Qwen Image：软糖动物LoRA", "description": "应用Qwen图像LoRA，将任何主题转化为酸软糖风格的角色。在提示中输入动物名称，生成具有风格化、可食用外观的生物。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图像", "LoRA" ], "models": [ "Qwen-Image" ], "date": "2026-03-30", "openSource": true, "size": 30709016166, "vram": 30709016166, "usage": 15, "searchRank": 0, "username": "enigmatic_e", "io": { "outputs": [ { "nodeId": 60, "nodeType": "SaveImage", "file": "template_sugar_coated_gummy_style_qwen.png", "mediaType": "image" } ] }, "thumbnail": [ "output/template_sugar_coated_gummy_style_qwen.png" ] }, { "name": "template_eric_thumbnail_generator", "title": "YouTube缩略图生成器", "description": "使用 Nano Banana Pro 生成高对比度、可点击的 YouTube 缩略图。通过夸张的主体和醒目的色彩策略，打造极具吸引力的戏剧性构图。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图像" ], "models": [ "Nano Banana Pro" ], "logos": [ { "provider": "Google" } ], "date": "2026-03-21", "openSource": false, "size": 0, "vram": 0, "usage": 26, "searchRank": 0, "username": "enigmatic_e", "io": { "inputs": [ { "nodeId": 16, "nodeType": "LoadImage", "file": "man_with_dog.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 9, "nodeType": "SaveImage", "file": "ComfyUI_00223_.png", "mediaType": "image" } ] }, "thumbnail": [ "output/ComfyUI_00223_.png" ] }, { "name": "templates_ohneis_i2v", "title": "图生视频", "description": "这是我Instagram上爆火的韦斯·安德森风格短视频背后的基础工作流程。它展示了如何从一个简单的提示词开始，通过电影级图像模型生成韦斯·安德森风格的图像，接着用Nanobanana增强纹理与细节，最后利用当前最佳视频模型Kling 3.0将其转化为视频。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "图生视频", "文生图", "视频" ], "models": [ "Kling 2.6", "Nano Banana Pro", "Z-Image-Turbo" ], "logos": [ { "provider": [ "Google", "Kling" ] } ], "date": "2026-03-19", "openSource": false, "size": 20723217203, "vram": 20723217203, "usage": 26, "searchRank": 0, "username": "ohneis652", "thumbnail": [ "thumbnail/templates_ohneis_i2v-1.png", "thumbnail/templates_ohneis_i2v-2.png" ] }, { "name": "templates_doc_workbox_poster_recreator", "title": "写实海报重设计，角色形象焕新", "description": "将上传的图片转化为逼真的海报设计。在保持艺术风格和配色方案的同时，改变人物姿势、面部和服装。输出具有艺术字体、视觉冲击力强的海报。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "海报", "图片编辑" ], "models": [ "Nano Banana Pro" ], "logos": [ { "provider": "Google" } ], "date": "2026-03-21", "openSource": false, "size": 0, "vram": 0, "usage": 18, "searchRank": 0, "username": "Doc_workBox", "io": { "inputs": [ { "nodeId": 3, "nodeType": "LoadImage", "file": "reference_poster.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 4, "nodeType": "SaveImage", "file": "templates_doc_workbox_poster_recreator.png", "mediaType": "image" } ] }, "thumbnail": [ "output/templates_doc_workbox_poster_recreator.png" ] }, { "name": "templates_rob_realistic_2k_images_quick_variations.app", "title": "生成逼真变体", "description": "上传参考图像，使用Grok生成描述并创建变体，通过快速2倍放大增加细节。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "图像", "图像放大" ], "models": [ "Grok", "Z-Image-Turbo" ], "logos": [ { "provider": "Grok" } ], "date": "2026-03-09", "openSource": false, "size": 20723217203, "vram": 20723217203, "usage": 63, "searchRank": 0, "io": { "inputs": [ { "nodeId": 77, "nodeType": "LoadImage", "file": "input_templates_rob_realistic_2k_images_quick_variations.png", "mediaType": "image" } ] }, "username": "hellorob", "thumbnail": [ "thumbnail/templates_rob_realistic_2k_images_quick_variations_app.mp4" ] }, { "name": "templates_rob_portrait_light_migration.app", "title": "人像重光", "description": "上传您的目标图像和参考光照图像。为目标图像重新打光。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "图片编辑", "LoRA", "重打光" ], "models": [ "Qwen-Image-Edit" ], "date": "2026-03-09", "openSource": true, "size": 31138512896, "vram": 31138512896, "usage": 41, "searchRank": 0, "io": { "inputs": [ { "nodeId": 6, "nodeType": "LoadImage", "file": "input_2-relight.png", "mediaType": "image" }, { "nodeId": 7, "nodeType": "LoadImage", "file": "input_1-relight.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 5, "nodeType": "SaveImage", "file": "templates_rob_portrait_light_migration_app.png", "mediaType": "image" } ] }, "username": "hellorob", "thumbnail": [ "input/input_1-relight.png", "output/templates_rob_portrait_light_migration_app.png" ] }, { "name": "template_sferro21_product_ad.app", "title": "产品广告电影序列生成器", "description": "上传产品图片和电影风格参考。生成一系列风格化产品广告，保持一致的构图和灯光。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "产品", "图片编辑", "替换", "电影感", "风格迁移", "样机" ], "models": [ "Nano Banana Pro" ], "logos": [ { "provider": "Google" } ], "date": "2026-03-16", "openSource": false, "size": 0, "vram": 0, "usage": 94, "searchRank": 0, "username": "sferro21", "io": { "inputs": [ { "nodeId": 1, "nodeType": "LoadImage", "file": "golf_ball_in_crocodile_mouth.png", "mediaType": "image" }, { "nodeId": 2, "nodeType": "LoadImage", "file": "comfy_hat.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 5, "nodeType": "SaveImage", "file": "template_sferro21_product_ad.app.png", "mediaType": "image" } ] }, "thumbnail": [ "thumbnail/template_sferro21_product_ad.png" ] }, { "name": "templates_3d_match_game_art_style.app", "title": "游戏资源风格迁移精灵图生成器", "description": "使用轮廓参考和风格参考作为输入，生成2D精灵作为模型纹理。适用于2D/2.5D游戏的背景布景。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "游戏", "风格参考" ], "models": [ "Nano Banana Pro" ], "logos": [ { "provider": "Google" } ], "date": "2026-03-06", "openSource": false, "size": 0, "vram": 0, "usage": 14, "searchRank": 0, "io": { "inputs": [ { "nodeId": 1, "nodeType": "LoadImage", "file": "game_style.png", "mediaType": "image" }, { "nodeId": 2, "nodeType": "LoadImage", "file": "gas_lamp.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 5, "nodeType": "SaveImage", "file": "templates_3d_match_game_art_style-1.png", "mediaType": "image" }, { "nodeId": 3, "nodeType": "SaveImage", "file": "templates_3d_match_game_art_style-2.png", "mediaType": "image" } ] }, "username": "shane", "thumbnail": [ "input/gas_lamp.png", "output/templates_3d_match_game_art_style-1.png" ] }, { "name": "template_sirolim_image_script_video", "title": "图像到脚本到Kling视频生成", "description": "上传图片以自动生成描述性脚本。该脚本随后将用于通过Kling 3.0 Omni模型创建视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生视频", "视频" ], "models": [ "Google Gemini", "Kling O3", "Nano Banana 2" ], "logos": [ { "provider": [ "Kling", "Google" ] } ], "date": "2026-03-09", "openSource": false, "size": 0, "vram": 0, "usage": 29, "searchRank": 0, "io": { "inputs": [ { "nodeId": 16, "nodeType": "LoadImage", "file": "input_storyboard_video.jpg", "mediaType": "image" } ], "outputs": [ { "nodeId": 7, "nodeType": "SaveVideo", "file": "output_image_script_video.mp4", "mediaType": "video" } ] }, "username": "sirolim", "thumbnail": [ "output/output_image_script_video.mp4" ] }, { "name": "template_rob_realistic_2k_images_quick_variations", "title": "逼真2K图像 - 快速变体", "description": "上传您想要生成变体的图像。工作流程将对输入图像进行描述，使用Grok Image生成，并通过Z-image进行超分辨率处理以添加逼真纹理。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "图像放大", "文生图" ], "models": [ "Grok", "Z-Image-Turbo" ], "logos": [ { "provider": "Grok" } ], "date": "2026-03-08", "openSource": false, "size": 20723217203, "vram": 20723217203, "usage": 63, "searchRank": 0, "io": { "inputs": [ { "nodeId": 77, "nodeType": "LoadImage", "file": "rapper_on_cliff_with_moon.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 103, "nodeType": "SaveImage", "file": "template_rob_realistic_2k_images_quick_variations.png", "mediaType": "image" } ] }, "username": "hellorob", "thumbnail": [ "input/rapper_on_cliff_with_moon.png", "output/template_rob_realistic_2k_images_quick_variations.png" ] }, { "name": "templates-character_sheet", "title": "360度角色旋转", "description": "上传角色，生成全身与特写360度旋转图。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "多角度" ], "models": [ "Gemini3 Pro Image Preview", "Google", "Nano Banana Pro" ], "date": "2026-01-13", "openSource": false, "usage": 91, "logos": [ { "provider": "Google" } ], "io": { "inputs": [ { "nodeId": 1, "nodeType": "LoadImage", "file": "templates-character_sheet_pink_silver_outfit.jpg", "mediaType": "image" } ] }, "size": 0, "vram": 0, "searchRank": 0, "username": "hellorob", "thumbnail": [ "thumbnail/templates-character_sheet.mp4" ] }, { "name": "templates-portrait_light_migration", "title": "角色肖像光线迁移", "description": "上传角色肖像和参考光线图像。生成应用参考光线并保持角色一致性的图像。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "LoRA", "肖像", "重打光" ], "models": [ "Qwen", "Qwen-Image-Edit" ], "date": "2026-01-13", "openSource": true, "usage": 20, "io": { "inputs": [ { "nodeId": 6, "nodeType": "LoadImage", "file": "woman_with_sunbeam.jpg", "mediaType": "image" }, { "nodeId": 7, "nodeType": "LoadImage", "file": "women_with_blue_orb.jpg", "mediaType": "image" } ] }, "username": "hellorob", "thumbnail": [ "thumbnail/templates-portrait_light_migration.mp4" ], "size": 31138512896, "vram": 31138512896, "searchRank": 0 }, { "name": "templates-image_to_real", "title": "插画转写实", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "description": "输入插画，生成超写实版本，使用Qwen Image Edit 2509。", "tags": [ "风格迁移" ], "models": [ "Qwen-Image-Edit" ], "date": "2025-12-20", "size": 0, "vram": 0, "usage": 281, "io": { "inputs": [ { "nodeId": 14, "nodeType": "LoadImage", "file": "robot_office_worker_in_space.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 18, "nodeType": "SaveImage", "file": "templates-image_to_real.png", "mediaType": "image" } ] }, "username": "hellorob", "openSource": true, "searchRank": 0, "thumbnail": [ "input/robot_office_worker_in_space.png", "output/templates-image_to_real.png" ] }, { "name": "template_eric_seedance_5_subject_and_outfit_combine", "title": "角色肖像的服装应用", "description": "上传您的角色肖像和服装参考。生成应用特定服装的新肖像，实现一致的衣柜控制。Nano Banana Pro 和 Nano Banana 2 的经济实惠替代方案。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "虚拟试穿" ], "models": [ "Seedream 4.5" ], "logos": [ { "provider": "ByteDance" } ], "date": "2026-03-08", "openSource": false, "size": 0, "vram": 0, "usage": 36, "searchRank": 0, "io": { "inputs": [ { "nodeId": 16, "nodeType": "LoadImage", "file": "subject_far.png", "mediaType": "image" }, { "nodeId": 26, "nodeType": "LoadImage", "file": "clothing.png", "mediaType": "image" }, { "nodeId": 226, "nodeType": "LoadImage", "file": "subject_close.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 11, "nodeType": "SaveImage", "file": "template_eric_seedance_5_subject_and_outfit_combine-2.png", "mediaType": "image" }, { "nodeId": 258, "nodeType": "SaveImage", "file": "template_eric_seedance_5_subject_and_outfit_combine-1.png", "mediaType": "image" } ] }, "username": "enigmatic_e", "thumbnail": [ "thumbnail/template_eric_seedance_5_subject_and_outfit_combine.png" ] }, { "name": "template_eric_exploded_view", "title": "爆炸视图动画", "description": "上传设备或物体的图像以生成起始和结束帧。工作流程将动画化过渡到爆炸视图并返回，输出无缝视频循环。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "首尾帧视频", "视频" ], "models": [ "Nano Banana 2", "Seedance 1.5 Pro" ], "logos": [ { "provider": [ "ByteDance", "Google" ] } ], "date": "2026-03-06", "openSource": false, "size": 0, "vram": 0, "usage": 9, "searchRank": 0, "io": { "inputs": [ { "nodeId": 29, "nodeType": "LoadImage", "file": "template_eric_exploded_view_first_frame.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 381, "nodeType": "SaveVideo", "file": "template_eric_exploded_view.mp4", "mediaType": "video" } ] }, "username": "enigmatic_e", "thumbnail": [ "output/template_eric_exploded_view.mp4" ] }, { "name": "template-recraft_create_style", "title": "风格迁移 - Recraft 文生图", "description": "上传1-5个风格参考并创建风格ID，以生成一致且符合品牌形象的图片。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "风格迁移" ], "models": [ "Recraft" ], "logos": [ { "provider": "Recraft" } ], "date": "2026-02-11", "openSource": false, "size": 0, "vram": 0, "usage": 13, "searchRank": 0, "io": { "inputs": [ { "nodeId": 1, "nodeType": "LoadImage", "file": "input_style_1.png", "mediaType": "image" }, { "nodeId": 23, "nodeType": "LoadImage", "file": "input_style_2.png", "mediaType": "image" }, { "nodeId": 24, "nodeType": "LoadImage", "file": "input_style_3.png", "mediaType": "image" }, { "nodeId": 25, "nodeType": "LoadImage", "file": "input_style_4.png", "mediaType": "image" }, { "nodeId": 26, "nodeType": "LoadImage", "file": "input_style_5.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 37, "nodeType": "SaveImage", "file": "template-recraft_create_style.png", "mediaType": "image" } ] }, "username": "hellorob", "thumbnail": [ "thumbnail/template-recraft_create_style.mp4" ] }, { "name": "templates-1_click_multiple_scene_angles-v1.0", "title": "一键多角度场景", "mediaType": "image", "mediaSubtype": "webp", "description": "上传场景图像，一键生成多视角。", "tags": [ "图片编辑" ], "models": [ "Qwen-Image-Edit" ], "date": "2025-12-08", "searchRank": 8, "size": 31198642438, "vram": 31198642438, "usage": 234, "io": { "inputs": [ { "nodeId": 25, "nodeType": "LoadImage", "file": "comfy_billboard.png", "mediaType": "image" } ] }, "username": "hellorob", "thumbnail": [ "thumbnail/templates-1_click_multiple_scene_angles-v1_0.png" ] }, { "name": "templates-1_input-multiple_styles_prompt.app", "title": "1个输入，多种提示风格", "description": "上传一张图片，15秒内生成12种风格化变体。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "图片编辑" ], "models": [ "Grok" ], "logos": [ { "provider": "Grok" } ], "date": "2026-02-19", "openSource": false, "size": 0, "vram": 0, "usage": 14, "searchRank": 0, "io": { "inputs": [ { "nodeId": 1, "nodeType": "LoadImage", "file": "lightning_man.png", "mediaType": "image" } ] }, "username": "hellorob", "thumbnail": [ "thumbnail/templates-1_input-multiple_styles_prompt_app.mp4" ] }, { "name": "templates-photo_to_product_vid", "title": "产品图片转视频：鞋", "mediaType": "image", "mediaSubtype": "webp", "description": "用手机拍照，上传生成专业级产品视频。", "tags": [ "产品", "图生视频", "API" ], "models": [ "Seedream", "MiniMax" ], "openSource": false, "date": "2025-12-18", "searchRank": 8, "size": 0, "vram": 0, "usage": 17, "io": { "inputs": [ { "nodeId": 1, "nodeType": "LoadImage", "file": "product_shoe.jpeg", "mediaType": "image" } ] }, "username": "hellorob", "thumbnail": [ "thumbnail/templates-photo_to_product_vid.mp4" ] }, { "name": "templates-6-key-frames", "title": "连续首尾帧视频拼接", "mediaType": "image", "mediaSubtype": "webp", "description": "连续6帧首尾视频拼接，动作补足", "tags": [ "首尾帧视频", "图生视频" ], "models": [ "Wan2.2" ], "date": "2025-12-03", "searchRank": 8, "size": 0, "vram": 0, "usage": 152, "io": { "inputs": [ { "nodeId": 62, "nodeType": "LoadImage", "file": "6-key-frames-1.jpg", "mediaType": "image" }, { "nodeId": 122, "nodeType": "LoadImage", "file": "6-key-frames-2.png", "mediaType": "image" }, { "nodeId": 124, "nodeType": "LoadImage", "file": "6-key-frames-3.png", "mediaType": "image" }, { "nodeId": 126, "nodeType": "LoadImage", "file": "6-key-frames-4.png", "mediaType": "image" }, { "nodeId": 128, "nodeType": "LoadImage", "file": "6-key-frames-5.png", "mediaType": "image" }, { "nodeId": 130, "nodeType": "LoadImage", "file": "6-key-frames-6.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "thumbnail/templates-6-key-frames.mp4" ], "openSource": true }, { "name": "templates-all_in_one-image_edit_models", "title": "1个输入与多个编辑模型对比", "description": "上传一张输入图像并选择多个编辑模型。生成并排输出以直接比较模型效果和质量。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "图片编辑" ], "models": [ "Flux.2 Dev", "Flux.2 Klein", "GPT-Image-1.5", "Grok", "Nano Banana Pro", "Qwen-Image-Edit", "Seedream" ], "logos": [ { "provider": [ "Google", "WAN", "OpenAI", "ByteDance", "Grok" ], "label": "Multiple Providers" } ], "date": "2026-02-18", "openSource": false, "size": 134002979635, "vram": 134002979635, "usage": 53, "searchRank": 0, "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadImage", "file": "little_bot.png", "mediaType": "image" } ] }, "username": "hellorob", "thumbnail": [ "thumbnail/templates-all_in_one-image_edit_models.mp4" ] }, { "name": "templates-subject_product_swap.app", "title": "交换手中商品，UGC风格", "mediaType": "image", "mediaSubtype": "webp", "description": "上传一张角色拿着产品和你品牌产品的照片，生成产品已互换的图片。", "tags": [ "API", "产品", "替换" ], "models": [ "Gemini3 Pro Image Preview", "Google", "Nano Banana Pro" ], "openSource": false, "date": "2025-12-18", "searchRank": 8, "size": 0, "vram": 0, "usage": 24, "logos": [ { "provider": "Google" } ], "io": { "inputs": [ { "nodeId": 10, "nodeType": "LoadImage", "file": "subject-girl-holding-skincare-bottle.png", "mediaType": "image" }, { "nodeId": 11, "nodeType": "LoadImage", "file": "product-cleanser.jpeg", "mediaType": "image" } ] }, "username": "hellorob", "thumbnail": [ "thumbnail/templates-subject_product_swap_app.png" ] }, { "name": "templates-product_scene_relight", "title": "产品场景合成与重新光照", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "description": "上传产品和背景图片，场景合成并自然重新打光，支持Seedream 4.5。", "tags": [ "图片编辑", "产品", "重打光" ], "models": [ "ByteDance", "Seedream" ], "openSource": false, "date": "2025-12-15", "searchRank": 8, "size": 0, "vram": 0, "usage": 36, "logos": [ { "provider": "ByteDance" } ], "io": { "inputs": [ { "nodeId": 1, "nodeType": "LoadImage", "file": "comfy_shampoo.png", "mediaType": "image" }, { "nodeId": 2, "nodeType": "LoadImage", "file": "flower_sea.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 23, "nodeType": "SaveImage", "file": "templates-product_scene_relight.png", "mediaType": "image" } ] }, "username": "hellorob", "thumbnail": [ "input/comfy_shampoo.png", "output/templates-product_scene_relight.png" ] }, { "name": "templates_mjm_looped_restyler", "title": "循环重制器", "description": "上传起始图像及其提示，以生成从第一帧到最后一帧的循环视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生视频", "视频", "首尾帧视频" ], "models": [ "Nano Banana", "Google Gemini", "Wan2.2" ], "logos": [ { "provider": [ "Google", "WAN" ] } ], "date": "2026-03-03", "openSource": false, "size": 38010460570, "vram": 38010460570, "usage": 6, "searchRank": 0, "io": { "inputs": [ { "nodeId": 4, "nodeType": "LoadImage", "file": "looped_restyler_init.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 80, "nodeType": "SaveVideo", "file": "templates_mjm_looped_restyler.mp4", "mediaType": "video" } ] }, "username": "Julien | MJM", "thumbnail": [ "output/templates_mjm_looped_restyler.mp4" ] }, { "name": "templates-3D_logo_texture_animation", "title": "动感3D标志动画", "mediaType": "image", "mediaSubtype": "webp", "description": "上传你的logo矢量图，设定想要的贴图，生成首尾帧，自动补全过程动画。", "tags": [ "品牌设计", "首尾帧视频" ], "models": [ "ByteDance", "Seedream", "Google Gemini", "Nano Banana Pro" ], "openSource": false, "date": "2025-12-15", "searchRank": 8, "size": 0, "vram": 0, "usage": 11, "logos": [ { "provider": [ "ByteDance", "Google" ] } ], "io": { "inputs": [ { "nodeId": 1, "nodeType": "LoadImage", "file": "comfy_logo_c.png", "mediaType": "image" } ] }, "username": "hellorob", "thumbnail": [ "thumbnail/templates-3D_logo_texture_animation.mp4" ] }, { "name": "templates-car_product", "title": "单图生成汽车视频", "mediaType": "image", "mediaSubtype": "webp", "description": "上传车辆照片，生成多角度高质量视频。", "tags": [ "产品", "图生视频", "API", "首尾帧视频" ], "models": [ "Seedream", "Kling" ], "openSource": false, "date": "2025-12-18", "searchRank": 8, "size": 0, "vram": 0, "usage": 25, "io": { "inputs": [ { "nodeId": 10, "nodeType": "LoadImage", "file": "product_car.jpg", "mediaType": "image" } ] }, "username": "hellorob", "thumbnail": [ "thumbnail/templates-car_product.mp4" ] }, { "name": "templates-textured_logo_elements", "title": "为 logo 添加纹理元素", "mediaType": "image", "mediaSubtype": "webp", "description": "上传你的 logo、纹理和元素。生成一个品牌化纹理 logo 的视频素材。", "tags": [ "品牌设计", "图生视频" ], "models": [ "Gemini3 Pro Image Preview", "Nano Banana Pro", "Google", "ByteDance", "Seedance" ], "openSource": false, "date": "2025-12-11", "searchRank": 8, "size": 0, "vram": 0, "usage": 26, "logos": [ { "provider": [ "ByteDance", "Google" ] } ], "io": { "inputs": [ { "nodeId": 4, "nodeType": "LoadImage", "file": "comfy_logo_b.png", "mediaType": "image" }, { "nodeId": 17, "nodeType": "LoadImage", "file": "furry_ball.png", "mediaType": "image" }, { "nodeId": 18, "nodeType": "LoadImage", "file": "furry_tech.png", "mediaType": "image" } ] }, "username": "hellorob", "thumbnail": [ "thumbnail/templates-textured_logo_elements.mp4" ] }, { "name": "templates-subject_holding_product.app", "title": "让模特手持产品", "mediaType": "image", "mediaSubtype": "webp", "description": "上传你的角色和产品照片，生成角色手持产品的图片。", "tags": [ "API", "肖像", "产品" ], "models": [ "Gemini3 Pro Image Preview", "Google", "Nano Banana Pro" ], "openSource": false, "date": "2025-12-18", "searchRank": 8, "size": 0, "vram": 0, "usage": 14, "logos": [ { "provider": "Google" } ], "io": { "inputs": [ { "nodeId": 10, "nodeType": "LoadImage", "file": "subject-elderly-lady-portrait.png", "mediaType": "image" }, { "nodeId": 11, "nodeType": "LoadImage", "file": "product-toothbrush.png", "mediaType": "image" } ] }, "username": "hellorob", "thumbnail": [ "thumbnail/templates-subject_holding_product_app.png" ] }, { "name": "templates-textured_logotype-v2.1", "title": "为 Logo 应用材质", "mediaType": "image", "mediaSubtype": "webp", "description": "上传你的 logo，添加材质和元素，打造品牌资产", "tags": [ "品牌设计", "图生视频", "首尾帧视频" ], "models": [ "Gemini3 Pro Image Preview", "Nano Banana Pro", "Google", "ByteDance", "Seedance" ], "date": "2025-12-03", "openSource": false, "searchRank": 8, "size": 0, "vram": 0, "usage": 5, "logos": [ { "provider": [ "ByteDance", "Google" ] } ], "io": { "inputs": [ { "nodeId": 4, "nodeType": "LoadImage", "file": "comfy_logo_b.png", "mediaType": "image" }, { "nodeId": 17, "nodeType": "LoadImage", "file": "logotype_texture.png", "mediaType": "image" } ] }, "username": "hellorob", "thumbnail": [ "thumbnail/templates-textured_logotype-v2_1.mp4" ] }, { "name": "templates-assemble_dieline", "title": "根据刀模生成品牌包装", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "hoverDissolve", "description": "上传你的产品刀模，组装成3D包装。", "tags": [ "图片编辑", "产品" ], "models": [ "Google", "Google Gemini Image", "Nano Banana Pro" ], "openSource": false, "date": "2025-12-15", "searchRank": 8, "size": 0, "vram": 0, "usage": 33, "logos": [ { "provider": "Google" } ], "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadImage", "file": "dieline.jpg", "mediaType": "image" } ], "outputs": [ { "nodeId": 6, "nodeType": "SaveImage", "file": "templates-assemble_dieline.png", "mediaType": "image" } ] }, "username": "hellorob", "thumbnail": [ "input/dieline.jpg", "output/templates-assemble_dieline.png" ] }, { "name": "templates-product_ad-v2.0", "title": "产品广告替换", "mediaType": "image", "mediaSubtype": "webp", "description": "以参考广告风格为你的产品创建静态广告", "tags": [ "风格参考" ], "models": [ "Gemini3 Pro Image Preview", "Nano Banana Pro", "Google", "ByteDance", "Seedance" ], "date": "2025-12-03", "openSource": false, "searchRank": 8, "size": 0, "vram": 0, "usage": 81, "logos": [ { "provider": [ "ByteDance", "Google" ] } ], "io": { "inputs": [ { "nodeId": 1, "nodeType": "LoadImage", "file": "reference_ad.png", "mediaType": "image" }, { "nodeId": 2, "nodeType": "LoadImage", "file": "comfy_gpu.png", "mediaType": "image" } ] }, "username": "hellorob", "thumbnail": [ "thumbnail/templates-product_ad-v2_0.png" ] }, { "name": "templates-color_illustration", "title": "线稿上色", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "description": "输入黑白线稿，为线稿填色。", "tags": [ "API" ], "models": [ "Gemini3 Pro Image Preview", "Google", "Nano Banana Pro" ], "openSource": false, "date": "2025-12-20", "searchRank": 8, "size": 0, "vram": 0, "usage": 2, "logos": [ { "provider": "Google" } ], "io": { "inputs": [ { "nodeId": 10, "nodeType": "LoadImage", "file": "illustration_bw.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 11, "nodeType": "SaveImage", "file": "templates-color_illustration.png", "mediaType": "image" } ] }, "username": "hellorob", "thumbnail": [ "input/illustration_bw.png", "output/templates-color_illustration.png" ] }, { "name": "template-multistyle-magazine-cover-nanobananapro", "title": "杂志封面与包装设计", "mediaType": "image", "mediaSubtype": "webp", "description": "为你的杂志封面照片设计文本排版，并探索其包装设计方案。", "tags": [ "图片编辑", "样机", "布局设计" ], "models": [ "Nano Banana Pro", "Google" ], "date": "2025-12-06", "searchRank": 8, "openSource": false, "size": 0, "vram": 0, "usage": 1, "logos": [ { "provider": "Google" } ], "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadImage", "file": "template-Magazine_Cover_&_Packag_Design.png", "mediaType": "image" } ] }, "username": "hellorob", "thumbnail": [ "thumbnail/template-multistyle-magazine-cover-nanobananapro.png" ] } ] }, { "moduleName": "default", "type": "image", "category": "生成类型", "icon": "icon-[lucide--image]", "title": "图像", "templates": [ { "name": "api_grok_image_edit", "title": "Grok：图像编辑", "description": "上传图片以使用Grok模型进行编辑。生成具有表现力、电影感视觉效果的修改版本", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "图片编辑" ], "models": [ "Grok" ], "date": "2026-01-28", "openSource": false, "usage": 2886, "io": { "inputs": [ { "nodeId": 6, "nodeType": "LoadImage", "file": "cyberpunk_robot.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 3, "nodeType": "SaveImage", "file": "api_grok_image_edit.png", "mediaType": "image" } ] }, "size": 0, "vram": 0, "searchRank": 0, "logos": [ { "provider": "Grok" } ], "username": "ComfyUI", "thumbnail": [ "input/cyberpunk_robot.png", "output/api_grok_image_edit.png" ] }, { "name": "api_grok_imagine_image_quality_image_edit", "title": "Grok Imagine 图像质量：编辑", "description": "通过Grok Imagine图像质量功能，利用先进的文本渲染和创意控制提升图像质量。输入图像或提示词以获得更佳的细节和保真度。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "图片编辑" ], "models": [ "Grok" ], "logos": [ { "provider": "Grok" } ], "date": "2026-05-06", "openSource": false, "size": 0, "vram": 0, "usage": 972, "searchRank": 0, "username": "ComfyUI" }, { "name": "image_z_image_turbo", "tutorialUrl": "https://docs.comfy.org/tutorials/image/z-image/z-image-turbo", "title": "Z-Image-Turbo 文生图", "mediaType": "image", "mediaSubtype": "webp", "description": "高效的图像生成基础模型，采用单流扩散变换器，支持中英文。", "tags": [ "图像", "文生图" ], "models": [ "Z-Image-Turbo" ], "date": "2025-11-27", "size": 20830591386, "vram": 20830591386, "usage": 4861, "searchRank": 0, "openSource": true, "io": { "outputs": [ { "nodeId": 9, "nodeType": "SaveImage", "file": "image_z_image_turbo.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "output/image_z_image_turbo.png" ] }, { "name": "image_qwen_image_edit_2509", "title": "Qwen Image Edit 2509", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "description": "高级图像编辑，支持多图像输入、改进的一致性和 ControlNet 集成。", "tutorialUrl": "https://docs.comfy.org/tutorials/image/qwen/qwen-image-edit", "tags": [ "图片编辑", "ControlNet" ], "models": [ "Qwen-Image" ], "date": "2025-09-25", "size": 31782757990, "vram": 31782757990, "usage": 3775, "io": { "inputs": [ { "nodeId": 78, "nodeType": "LoadImage", "file": "image_qwen_image_edit_2509_input_image.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 60, "nodeType": "SaveImage", "file": "image_qwen_image_edit_2509.png", "mediaType": "image" }, { "nodeId": 342, "nodeType": "SaveImage", "file": "image_qwen_image_edit_2509.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "input/image_qwen_image_edit_2509_input_image.png", "output/image_qwen_image_edit_2509.png" ], "openSource": true, "searchRank": 0 }, { "name": "api_nano_banana_pro", "title": "Nano Banana Pro", "description": "Nano-banana Pro (Gemini 3.0 Pro Image) - 工作室级 4K 图像生成和编辑，增强文本渲染和角色一致性。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "hoverDissolve", "tags": [ "API", "图像", "图片编辑" ], "models": [ "Gemini3 Pro Image Preview", "Google", "Nano Banana Pro" ], "date": "2025-11-21", "openSource": false, "size": 0, "vram": 0, "usage": 3519, "io": { "inputs": [ { "nodeId": 11, "nodeType": "LoadImage", "file": "api_nano_banana_pro_input_image_1.png", "mediaType": "image" }, { "nodeId": 12, "nodeType": "LoadImage", "file": "api_nano_banana_pro_input_image_2.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 30, "nodeType": "SaveImage", "file": "api_nano_banana_pro.png", "mediaType": "image" } ] }, "logos": [ { "provider": "Google" } ], "searchRank": 0, "username": "ComfyUI", "thumbnail": [ "output/api_nano_banana_pro.png" ] }, { "name": "api_bytedance_seedream4", "title": "字节跳动 Seedream 4.0", "description": "多模态 AI 模型，支持文生图和图像编辑。使用自然语言控制，在 2 秒内生成 2K 图像。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "图像", "图片编辑", "文生图" ], "models": [ "ByteDance", "Seedream 4.0" ], "date": "2025-09-11", "openSource": false, "size": 0, "vram": 0, "usage": 2027, "io": { "inputs": [ { "nodeId": 11, "nodeType": "LoadImage", "file": "comfyui_logo.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 12, "nodeType": "SaveImage", "file": "api_bytedance_seedream4.png", "mediaType": "image" } ] }, "logos": [ { "provider": "ByteDance" } ], "searchRank": 0, "username": "ComfyUI", "thumbnail": [ "output/api_bytedance_seedream4.png" ] }, { "name": "api_grok_text_to_image", "title": "Grok：文生图", "description": "输入文本提示，使用Grok模型快速生成高质量图像。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "文生图" ], "models": [ "Grok" ], "date": "2026-01-28", "openSource": false, "usage": 404, "size": 0, "vram": 0, "searchRank": 0, "logos": [ { "provider": "Grok" } ], "io": { "outputs": [ { "nodeId": 3, "nodeType": "SaveImage", "file": "api_grok_text_to_image.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "output/api_grok_text_to_image.png" ] }, { "name": "api_google_nano_banana2_image_edit", "title": "Nano Banana 2：图像编辑", "description": "上传一张图片和一个文本提示来编辑它。根据您的指令生成一张修改原图的新图片，保持主体一致性和高视觉保真度。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "图片编辑" ], "models": [ "Google", "Nano Banana 2" ], "logos": [ { "provider": "Google" } ], "date": "2026-02-26", "openSource": false, "size": 0, "vram": 0, "usage": 2120, "searchRank": 0, "io": { "inputs": [ { "nodeId": 16, "nodeType": "LoadImage", "file": "man_with_dog.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 9, "nodeType": "SaveImage", "file": "api_google_nano_banana2_image_edit.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "image_qwen_image_edit_2511", "tutorialUrl": "https://docs.comfy.org/tutorials/image/qwen/qwen-image-edit-2511", "title": "Qwen Image Edit 2511 - 材质替换", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "description": "通过组合参考图像，使用 Qwen-Image-Edit-2511 替换物体（如家具）的材质。", "tags": [ "图片编辑" ], "models": [ "Qwen-Image-Edit" ], "date": "2025-12-23", "size": 51324859187, "vram": 51324859187, "usage": 1238, "io": { "inputs": [ { "nodeId": 41, "nodeType": "LoadImage", "file": "leather_sofa.png", "mediaType": "image" }, { "nodeId": 83, "nodeType": "LoadImage", "file": "texture_fur.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 9, "nodeType": "SaveImage", "file": "image_qwen_image_edit_2511.png", "mediaType": "image" } ] }, "openSource": true, "searchRank": 0, "username": "ComfyUI", "thumbnail": [ "input/leather_sofa.png", "output/image_qwen_image_edit_2511.png" ] }, { "name": "api_krea2_t2i", "title": "Krea2：文生图", "description": "输入描述您所需场景的文字提示。生成四张美观、高质量的图像，以创造性的多样性诠释您的提示。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "文生图" ], "models": [ "Krea 2" ], "logos": [ { "provider": "Krea" } ], "date": "2026-05-27", "openSource": false, "size": 0, "vram": 0, "usage": 962, "searchRank": 0, "username": "ComfyUI", "io": { "outputs": [ { "nodeId": 2, "nodeType": "SaveImage", "file": "api_krea2_t2i.png", "mediaType": "image" } ] }, "thumbnail": [ "output/api_krea2_t2i.png" ] }, { "name": "api_bytedance_seedream_5_0_lite_image_edit", "title": "Seedream 5.0 Lite：图像编辑", "description": "上传一张图片和一个文本提示来编辑它。生成一张修改后的图片，其中的细节与您的指令保持一致。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "图片编辑" ], "models": [ "Seedream 5.0 Lite" ], "logos": [ { "provider": "ByteDance" } ], "date": "2026-02-24", "openSource": false, "size": 0, "vram": 0, "usage": 1565, "searchRank": 0, "io": { "inputs": [ { "nodeId": 29, "nodeType": "LoadImage", "file": "man_with_old_computers.png", "mediaType": "image" }, { "nodeId": 32, "nodeType": "LoadImage", "file": "inflated_character.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 26, "nodeType": "SaveImage", "file": "api_bytedance_seedream_5_0_lite_image_edit.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "input/man_with_old_computers.png", "output/api_bytedance_seedream_5_0_lite_image_edit.png" ] }, { "name": "api_openai_gpt_image_2_image_edit", "title": "GPT图像2：图像编辑", "description": "上传参考图像并编写提示词，即可创建适合营销活动的海报、产品照片和多语言标识。该模型可生成清晰的文字和逼真的输出，适用于商业用途。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图片编辑", "API" ], "models": [ "GPT Image 2" ], "logos": [ { "provider": "OpenAI" } ], "date": "2026-04-22", "openSource": false, "size": 0, "vram": 0, "usage": 1159, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 271, "nodeType": "LoadImage", "file": "product_shoe.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 269, "nodeType": "SaveImage", "file": "api_openai_gpt_image_2_image_edit.png", "mediaType": "image" } ] }, "thumbnail": [ "output/api_openai_gpt_image_2_image_edit.png" ] }, { "name": "api_openai_gpt_image_2_t2i", "title": "GPT图像2：文本转图像", "description": "使用OpenAI最新模型创建可用于宣传活动的海报、产品照片和多语言标牌。上传可选的参考图像，编写详细的提示词，生成高质量、逼真的输出。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生图" ], "models": [ "GPT Image 2" ], "logos": [ { "provider": "OpenAI" } ], "date": "2026-04-22", "openSource": false, "size": 0, "vram": 0, "usage": 1060, "searchRank": 0, "username": "ComfyUI", "io": { "outputs": [ { "nodeId": 269, "nodeType": "SaveImage", "file": "api_openai_gpt_image_2_t2i.png", "mediaType": "image" } ] }, "thumbnail": [ "output/api_openai_gpt_image_2_t2i.png" ] }, { "name": "image_flux2", "tutorialUrl": "https://docs.comfy.org/tutorials/flux/flux-2-dev", "title": "Flux.2 Dev", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "description": "生成具有多参考一致性和专业文本渲染的写实照片级图像。", "tags": [ "图像", "图片编辑", "文生图" ], "models": [ "Flux", "Flux.2 Dev" ], "date": "2025-11-26", "size": 74088185856, "vram": 74088185856, "usage": 1711, "io": { "inputs": [ { "nodeId": 46, "nodeType": "LoadImage", "file": "image_flux2_input_image.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 9, "nodeType": "SaveImage", "file": "image_flux2.png", "mediaType": "image" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0, "thumbnail": [ "input/image_flux2_input_image.png", "output/image_flux2.png" ] }, { "name": "image_z_image", "title": "Z-Image: 文生图", "mediaType": "image", "mediaSubtype": "webp", "description": "创意自由的基础。多样美学与出色写实品质；适合微调；对负面提示敏感；生成多样性高。", "tags": [ "图像", "文生图" ], "models": [ "Z-Image" ], "date": "2026-01-27", "size": 20830591386, "vram": 20830591386, "usage": 772, "tutorialUrl": "https://docs.comfy.org/tutorials/image/z-image/z-image", "searchRank": 0, "openSource": true, "io": { "outputs": [ { "nodeId": 9, "nodeType": "SaveImage", "file": "image_z_image.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "output/image_z_image.png" ] }, { "name": "image_qwen_Image_2512", "tutorialUrl": "https://docs.comfy.org/tutorials/image/qwen/qwen-image-2512", "title": "Qwen Image 2512", "mediaType": "image", "mediaSubtype": "webp", "description": "文本到图像模型，具备更真实的人体细节、更精细的自然风景和动物毛发效果，以及更准确的文字排版和多模态融合。", "tags": [ "图像", "文生图" ], "models": [ "Qwen", "Qwen-Image" ], "date": "2025-12-31", "size": 31782757990, "vram": 31782757990, "usage": 573, "searchRank": 0, "openSource": true, "io": { "outputs": [ { "nodeId": 60, "nodeType": "SaveImage", "file": "image_qwen_Image_2512.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "output/image_qwen_Image_2512.png" ] }, { "name": "image_firered_image_edit1_1", "title": "FireRed 图像编辑 1.1：图像编辑", "description": "上传一张图片和一个文本提示来编辑它。生成一张经过修改的图片，具有增强的身份一致性、多元素融合或专业照片修复功能。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "图片编辑" ], "models": [ "FireRed Image Edit 1.1" ], "date": "2026-03-06", "openSource": true, "size": 51324859187, "vram": 51324859187, "usage": 1244, "searchRank": 0, "io": { "inputs": [ { "nodeId": 143, "nodeType": "LoadImage", "file": "future_portrait.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 9, "nodeType": "SaveImage", "file": "image_firered_image_edit1_1.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "input/future_portrait.png", "output/image_firered_image_edit1_1.png" ] }, { "name": "api_grok_imagine_image_quality_image_generation", "title": "Grok Imagine Image Quality: 生成", "description": "使用 Grok Imagine 中的高级质量模式，生成细节更丰富、文本渲染更佳且创意控制更强的增强图像。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "图像" ], "models": [ "Grok" ], "logos": [ { "provider": "Grok" } ], "date": "2026-05-06", "openSource": false, "size": 0, "vram": 0, "usage": 198, "searchRank": 0, "username": "ComfyUI", "io": { "outputs": [ { "nodeId": 3, "nodeType": "SaveImage", "file": "api_grok_imagine_image_quality_image_generation.png", "mediaType": "image" } ] } }, { "name": "api_bytedance_seedream_5_0_lite_t2i", "title": "Seedream 5.0 Lite：文生图", "description": "输入您的文本提示以生成详细图像。此工作流程根据您的描述创建视觉呈现，精确控制样式和布局。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "文生图" ], "models": [ "Seedream 5.0 Lite" ], "logos": [ { "provider": "ByteDance" } ], "date": "2026-02-24", "openSource": false, "size": 0, "vram": 0, "usage": 168, "searchRank": 0, "io": { "outputs": [ { "nodeId": 26, "nodeType": "SaveImage", "file": "api_bytedance_seedream_5_0_lite_t2i.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "output/api_bytedance_seedream_5_0_lite_t2i.png" ] }, { "name": "api_luma_uni_1_image_create", "title": "Luma UNI-1：图像创作", "description": "使用Luma UNI-1节点，结合最多9张参考图像，生成全新的原创图像，涵盖风格、角色、色彩、光照和构图。通过详细提示、可调宽高比和种子控制，创建独特的构图。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "图像", "风格参考" ], "models": [ "Luma UNI-1" ], "logos": [ { "provider": "Luma" } ], "date": "2026-05-05", "openSource": false, "size": 0, "vram": 0, "usage": 124, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 33, "nodeType": "LoadImage", "file": "uni_1_style_ref.png", "mediaType": "image" } ] } }, { "name": "api_luma_uni_1_image_edit", "title": "Luma UNI-1：图像编辑", "description": "使用Uni-1的图像编辑模式编辑现有图像，保留构图和核心元素。支持通过提示和参考图像进行有针对性的调整，如光照和颜色。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "图片编辑", "API" ], "models": [ "Luma UNI-1" ], "logos": [ { "provider": "Luma" } ], "date": "2026-05-05", "openSource": false, "size": 0, "vram": 0, "usage": 186, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 33, "nodeType": "LoadImage", "file": "the_leap.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 34, "nodeType": "SaveImage", "file": "api_luma_uni_1_image_edit.png", "mediaType": "image" } ] } }, { "name": "image_hidream_o1_dev", "tutorialUrl": "https://docs.comfy.org/tutorials/image/hidream/hidream-o1", "title": "HiDream O1 Dev", "description": "输入文本提示和可选的参考图像。生成高分辨率图像（最高2048x2048），支持文本到图像、图像编辑和主题驱动的个性化。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "图片编辑", "文生图" ], "models": [ "HiDream" ], "date": "2026-05-14", "openSource": true, "size": 17072495002, "vram": 17072495002, "usage": 518, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 213, "nodeType": "LoadImage", "file": "noir_portrait.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 227, "nodeType": "SaveImage", "file": "image_hidream_o1_dev.png", "mediaType": "image" } ] }, "thumbnail": [ "input/noir_portrait.png", "output/image_hidream_o1_dev.png" ] }, { "name": "image_ernie_image_turbo", "tutorialUrl": "https://docs.comfy.org/tutorials/image/ernie-image/ernie-image", "title": "Ernie Image Turbo：文生图", "description": "使用ERNIE-Image turbo模型从文本提示生成图像。输入文本描述，即可获得具有精确文本渲染和结构化布局的高质量图像。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生图", "Turbo" ], "models": [ "ERNIE-Image" ], "date": "2026-04-14", "openSource": true, "size": 25984552141, "vram": 25984552141, "usage": 205, "searchRank": 0, "username": "ComfyUI", "io": { "outputs": [ { "nodeId": 73, "nodeType": "SaveImage", "file": "image_ernie_image_turbo.png", "mediaType": "image" } ] }, "thumbnail": [ "output/image_ernie_image_turbo.png" ] }, { "name": "image_hidream_o1", "tutorialUrl": "https://docs.comfy.org/tutorials/image/hidream/hidream-o1", "title": "HiDream O1 Full：图像生成", "description": "输入文本提示词，并可选择上传参考图像。通过文本到图像、编辑或主题驱动个性化，生成高达2048x2048的高分辨率图像。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生图", "图片编辑" ], "models": [ "HiDream" ], "date": "2026-05-14", "openSource": true, "size": 25447681229, "vram": 25447681229, "usage": 313, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 213, "nodeType": "LoadImage", "file": "noir_portrait.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 227, "nodeType": "SaveImage", "file": "image_hidream_o1.png", "mediaType": "image" } ] }, "thumbnail": [ "output/image_hidream_o1.png" ] }, { "name": "api_kling_o3_image", "title": "Kling O3：图像生成", "description": "上传最多10张参考图像以指导合成。使用Kling O3模型，基于您提供的视觉输入生成新图像。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "图片编辑", "文生图" ], "models": [ "Kling O3" ], "logos": [ { "provider": "Kling" } ], "date": "2026-02-10", "openSource": false, "size": 0, "vram": 0, "usage": 209, "searchRank": 0, "io": { "inputs": [ { "nodeId": 31, "nodeType": "LoadImage", "file": "vintage_portrait.png", "mediaType": "image" }, { "nodeId": 32, "nodeType": "LoadImage", "file": "Urban_Fashion.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 29, "nodeType": "SaveImage", "file": "api_kling_o3_image.json", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "output/api_kling_o3_image.json" ] }, { "name": "image_ideogram4_t2i", "title": "Ideogram v4：文本转图像", "description": "输入文本提示或结构化JSON描述。使用Ideogram 4.0生成具有精确布局、颜色和风格控制的图像。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生图" ], "models": [ "Ideogram" ], "date": "2026-06-03", "openSource": true, "size": 26521423053, "vram": 26521423053, "usage": 0, "searchRank": 0, "username": "ComfyUI" }, { "name": "api_ideogram_v4_t2i", "title": "Ideogram v4：文生图（API）", "description": "以结构化JSON格式输入文本提示。生成具有精确布局控制、文本渲染和调色板支持的高质量图像。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "文生图" ], "models": [ "Ideogram" ], "logos": [ { "provider": "Ideogram" } ], "date": "2026-06-03", "openSource": false, "size": 0, "vram": 0, "usage": 0, "searchRank": 0, "username": "ComfyUI" }, { "name": "image_flux2_klein_9b_kv_image_edit", "title": "Flux.2 Klein KV：图像编辑", "description": "上传参考图像和文本提示。使用缓存的图像数据生成多个编辑后的图像，以加快处理速度。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "图像", "图片编辑" ], "models": [ "Flux.2 Klein" ], "date": "2026-03-12", "openSource": true, "size": 18790481920, "vram": 18790481920, "usage": 16, "searchRank": 0, "io": { "inputs": [ { "nodeId": 76, "nodeType": "LoadImage", "file": "man_in_street.png", "mediaType": "image" }, { "nodeId": 81, "nodeType": "LoadImage", "file": "safari_outfit.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 94, "nodeType": "SaveImage", "file": "image_flux2_klein_9b_kv_image_edit.png", "mediaType": "image" } ] }, "username": "ComfyUI", "includeOnDistributions": [ "local" ], "thumbnail": [ "input/man_in_street.png", "output/image_flux2_klein_9b_kv_image_edit.png" ] }, { "name": "image_ernie_image", "tutorialUrl": "https://docs.comfy.org/tutorials/image/ernie-image/ernie-image", "title": "Ernie Image：文生图", "description": "使用ERNIE-Image模型从文本提示生成图像。输入文本描述，即可生成风格多样、细节丰富的结构化视觉效果。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生图" ], "models": [ "ERNIE-Image" ], "date": "2026-04-14", "openSource": true, "size": 25984552141, "vram": 25984552141, "usage": 107, "searchRank": 0, "username": "ComfyUI", "io": { "outputs": [ { "nodeId": 73, "nodeType": "SaveImage", "file": "image_ernie_image.png", "mediaType": "image" } ] }, "thumbnail": [ "output/image_ernie_image.png" ] }, { "name": "api_google_gemini_image", "title": "Google Gemini 图像", "description": "Nano-banana (Gemini-2.5-Flash Image) - 使用一致性进行图像编辑。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图片编辑", "图像", "API", "文生图" ], "models": [ "Gemini-2.5-Flash", "Nano Banana", "Google" ], "date": "2025-08-27", "openSource": false, "size": 0, "vram": 0, "usage": 250, "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadImage", "file": "comfyui_logo.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_recraft_v4_t2i", "title": "Recraft v4：文生图", "description": "输入文本提示以生成高分辨率图像。使用Recraft V4模型以获得增强的真实感、文本准确性和详细的矢量图形。根据您选择的版本输出1024x1024或2048x2048的图像。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "图像", "文生图" ], "models": [ "Recraft" ], "logos": [ { "provider": "Recraft" } ], "date": "2026-02-17", "openSource": false, "size": 0, "vram": 0, "usage": 103, "searchRank": 0, "io": { "outputs": [ { "nodeId": 5, "nodeType": "SaveImage", "file": "api_recraft_v4_t2i.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "output/api_recraft_v4_t2i.png" ] }, { "name": "image_qwen_Image_2512_controlnet", "tutorialUrl": "https://docs.comfy.org/tutorials/image/qwen/qwen-image-2512", "title": "Qwen-Image 2512：趣味联盟控制网", "description": "上传一张图片，并从Canny、HED、Depth、Pose、MLSD、Scribble或Grayscale中选择一种控制类型。根据所选的结构条件生成一张新图像。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "ControlNet", "图像" ], "models": [ "Qwen-Image 2512" ], "date": "2026-02-16", "openSource": true, "size": 35326106010, "vram": 35326106010, "usage": 191, "searchRank": 0, "io": { "inputs": [ { "nodeId": 121, "nodeType": "LoadImage", "file": "low_angle_shot.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 60, "nodeType": "SaveImage", "file": "image_qwen_Image_2512_controlnet.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "image_longcat_text_to_image", "title": "LongCat Image：文生图", "description": "根据文本提示生成图像。输入您想要的场景描述（英文或中文）。输出具有准确文本渲染的逼真图像。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图像", "文生图" ], "models": [ "LongCat" ], "date": "2026-03-02", "openSource": true, "size": 22226455757, "vram": 22226455757, "usage": 72, "searchRank": 0, "io": { "outputs": [ { "nodeId": 9, "nodeType": "SaveImage", "file": "image_longcat_text_to_image.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "output/image_longcat_text_to_image.png" ] }, { "name": "api_bfl_flux2_max_sofa_swap", "title": "BFL FLUX.2 [max]: 物体替换", "description": "使用FLUX.2 [max]实现高质量物体替换，适用于产品拍摄、家居更换，轻松保持场景一致性。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "图像", "图片编辑" ], "models": [ "Flux", "Flux.2" ], "date": "2025-12-22", "searchRank": 7, "openSource": false, "size": 0, "vram": 0, "usage": 89, "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadImage", "file": "minimalist_wooden_sofa_living_room.png", "mediaType": "image" }, { "nodeId": 3, "nodeType": "LoadImage", "file": "fluffy_beige_sofa.png", "mediaType": "image" }, { "nodeId": 10, "nodeType": "LoadImage", "file": "black_coat_model.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 6, "nodeType": "SaveImage", "file": "api_bfl_flux2_max_sofa_swap.png", "mediaType": "image" } ] }, "username": "ComfyUI", "logos": [ { "provider": "Black Forest Labs" } ], "thumbnail": [ "output/api_bfl_flux2_max_sofa_swap.png" ] }, { "name": "image_anima_base_v1", "tutorialUrl": "https://docs.comfy.org/tutorials/image/anima/anima", "title": "Anima Base v1：文本到图像", "description": "Input a text prompt describing an anime or artistic illustration. Generate a non-photorealistic image focused on anime concepts, characters, or styles.", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "动漫", "文生图" ], "models": [ "Anima" ], "date": "2026-05-23", "openSource": true, "size": 5583457485, "vram": 5583457485, "usage": 881, "searchRank": 0, "username": "ComfyUI", "io": { "outputs": [ { "nodeId": 46, "nodeType": "SaveImage", "file": "image_anima_base_v1.png", "mediaType": "image" } ] }, "thumbnail": [ "output/image_anima_base_v1.png" ] }, { "name": "api_kling_v3_t2i", "title": "Kling 3.0：文生图", "description": "使用 Kling 3.0 模型根据文本提示生成图像。输入您的描述性文本，即可获得高质量、与提示高度吻合的视觉输出。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "图像" ], "models": [ "Kling", "Kling 3.0" ], "logos": [ { "provider": "Kling" } ], "date": "2026-02-10", "openSource": false, "size": 0, "vram": 0, "usage": 112, "searchRank": 0, "io": { "outputs": [ { "nodeId": 29, "nodeType": "SaveImage", "file": "api_kling_v3_t2i.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "output/api_kling_v3_t2i.png" ] }, { "name": "image_flux2_klein_image_edit_4b_base", "title": "Flux.2 [Klein] 4B：图片编辑", "mediaType": "image", "mediaSubtype": "webp", "description": "更小型的基础模型，具备卓越的模型大小与质量比。非常适合本地部署、在有限硬件上微调，以及高效生成与编辑工作流。", "tags": [ "图像", "图片编辑" ], "models": [ "Flux", "Flux.2 Klein" ], "date": "2026-01-16", "size": 17072495002, "vram": 9878424781, "usage": 418, "tutorialUrl": "https://docs.comfy.org/tutorials/flux/flux-2-klein", "searchRank": 0, "openSource": true, "io": { "inputs": [ { "nodeId": 76, "nodeType": "LoadImage", "file": "robed_women.png", "mediaType": "image" }, { "nodeId": 81, "nodeType": "LoadImage", "file": "pink_tone_chair.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "thumbnail/image_flux2_klein_image_edit_4b_base.png" ] }, { "name": "image_flux2_klein_image_edit_9b_base", "title": "Flux.2 [Klein] 9B：图像编辑", "mediaType": "image", "mediaSubtype": "webp", "description": "BFL 未蒸馏基础模型。灵活性与可控性最高。非常适合微调。", "tags": [ "图像", "文生图" ], "models": [ "Flux", "Flux.2 Klein" ], "date": "2026-01-16", "size": 19327352832, "vram": 23300197581, "usage": 1234, "openSource": true, "searchRank": 0, "tutorialUrl": "https://docs.comfy.org/tutorials/flux/flux-2-klein", "io": { "inputs": [ { "nodeId": 76, "nodeType": "LoadImage", "file": "car_interior_white.jpeg", "mediaType": "image" }, { "nodeId": 81, "nodeType": "LoadImage", "file": "comfy_logo_blue.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "thumbnail/image_flux2_klein_image_edit_9b_base.png" ] }, { "name": "image_flux2_text_to_image_9b", "tutorialUrl": "https://docs.comfy.org/tutorials/flux/flux-2-klein", "title": "Flux.2 [Klein] 9B：文生图", "mediaType": "image", "mediaSubtype": "webp", "description": "BFL 蒸馏模型。亚秒级速度下仍具出色画质。适合实时生成并保留质量。营销发布将重点聚焦该模型。", "tags": [ "图像", "文生图" ], "models": [ "Flux", "Flux.2 Klein" ], "date": "2026-01-16", "size": 19219978650, "vram": 19219978650, "usage": 393, "openSource": true, "searchRank": 0, "username": "ComfyUI", "thumbnail": [ "output/image_flux2_text_to_image_9b.png" ] }, { "name": "image_flux2_text_to_image", "tutorialUrl": "https://docs.comfy.org/tutorials/flux/flux-2-dev", "title": "Flux.2 Dev 文生图", "mediaType": "image", "mediaSubtype": "webp", "description": "具备增强光效、材质和真实细节图像生成。", "tags": [ "图像", "文生图" ], "models": [ "Flux", "Flux.2 Dev" ], "date": "2025-11-26", "size": 71403831296, "vram": 71403831296, "usage": 515, "io": { "outputs": [ { "nodeId": 9, "nodeType": "SaveImage", "file": "image_flux2_text_to_image.png", "mediaType": "image" } ] }, "openSource": true, "searchRank": 0, "thumbnail": [ "output/image_flux2_text_to_image.png" ], "username": "ComfyUI" }, { "name": "image_flux2_klein_image_edit_9b_distilled", "tutorialUrl": "https://docs.comfy.org/tutorials/flux/flux-2-klein", "title": "Flux.2 [Klein] 9B 蒸馏：图像编辑", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "description": "BFL 蒸馏模型。亚秒级速度下仍具出色画质。适合实时生成并保留质量。营销发布将重点聚焦该模型。", "tags": [ "图像", "文生图" ], "models": [ "Flux", "Flux.2 Klein" ], "date": "2026-01-16", "size": 19112604467, "vram": 21045339750, "usage": 2, "searchRank": 0, "openSource": true, "io": { "inputs": [ { "nodeId": 76, "nodeType": "LoadImage", "file": "bold_outfit_woman.jpeg", "mediaType": "image" }, { "nodeId": 121, "nodeType": "LoadImage", "file": "handbag_white.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 9, "nodeType": "SaveImage", "file": "image_flux2_klein_image_edit_9b_distilled.png", "mediaType": "image" } ] }, "username": "ComfyUI", "includeOnDistributions": [ "local" ], "thumbnail": [ "output/image_flux2_klein_image_edit_9b_distilled.png", "output/image_flux2_klein_image_edit_9b_distilled.png" ] }, { "name": "image-qwen_image_edit_2511_lora_inflation", "title": "使用 INFL8 LoRA 的图像膨胀效果", "description": "上传一张图片并指定一个主体。使用INFL8 LoRA for Qwen Image Edit生成该主体的膨胀版本。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "图像" ], "models": [ "Qwen-Image-Edit" ], "date": "2026-01-30", "openSource": true, "size": 50787988275, "vram": 50787988275, "usage": 25, "searchRank": 0, "io": { "inputs": [ { "nodeId": 41, "nodeType": "LoadImage", "file": "inflation_lora_input.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 9, "nodeType": "SaveImage", "file": "image-qwen_image_edit_2511_lora_inflation.png", "mediaType": "image" } ] }, "username": "PurzBeats", "thumbnail": [ "input/inflation_lora_input.png", "output/image-qwen_image_edit_2511_lora_inflation.png" ] }, { "name": "api_recraft_style_reference", "title": "Recraft 风格参考生成", "description": "上传多张图片以创建一致的风格参考。使用此参考生成新图像，以保持输出之间的视觉连贯性。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "图像", "风格参考" ], "models": [ "Recraft" ], "logos": [ { "provider": "Recraft" } ], "date": "2026-02-05", "openSource": false, "size": 0, "vram": 0, "usage": 32, "searchRank": 0, "io": { "inputs": [ { "nodeId": 29, "nodeType": "LoadImage", "file": "recraft_style_ref-1.png", "mediaType": "image" }, { "nodeId": 36, "nodeType": "LoadImage", "file": "recraft_style_ref-3.png", "mediaType": "image" }, { "nodeId": 38, "nodeType": "LoadImage", "file": "recraft_style_ref-2.png", "mediaType": "image" }, { "nodeId": 39, "nodeType": "LoadImage", "file": "recraft_style_ref-4.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 34, "nodeType": "SaveImage", "file": "api_recraft_style_reference.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "output/api_recraft_style_reference.png" ] }, { "name": "api_recraft_v4_text_to_vector", "title": "Recraft v4：文生矢量", "description": "输入文本提示以生成干净、可编辑的矢量SVG图形。创建适用于设计工具的结构化路径的徽标、图标或插图。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "图像", "矢量" ], "models": [ "Recraft" ], "logos": [ { "provider": "Recraft" } ], "date": "2026-02-17", "openSource": false, "size": 0, "vram": 0, "usage": 15, "searchRank": 0, "username": "ComfyUI" }, { "name": "image_qwen_image_2512_with_2steps_lora", "tutorialUrl": "https://docs.comfy.org/tutorials/image/qwen/qwen-image-2512", "title": "Qwen-Image 2512 Turbo", "description": "Qwen-Image 2512 图像生成采用 2 步 Turbo LoRA 技术。输入您的文本提示即可快速生成结果，但为了速度会牺牲一些质量。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图像", "文生图", "Turbo" ], "models": [ "Qwen-Image 2512" ], "date": "2026-01-30", "openSource": true, "usage": 206, "username": "ComfyUI", "size": 32427003085, "vram": 32427003085, "searchRank": 0, "io": { "outputs": [ { "nodeId": 123, "nodeType": "SaveImage", "file": "image_qwen_image_2512_with_2steps_lora.png", "mediaType": "image" } ] }, "thumbnail": [ "output/image_qwen_image_2512_with_2steps_lora.png" ] }, { "name": "image_z_image_turbo_fun_union_controlnet", "tutorialUrl": "https://docs.comfy.org/tutorials/image/z-image/z-image-turbo", "title": "Z-Image-Turbo 控制生图", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "description": "用于 Z-Image-Turbo，支持 Canny、HED、深度、姿态、MLSD 等多种控制类型的 ControlNet。", "tags": [ "图像", "ControlNet" ], "models": [ "Z-Image-Turbo" ], "date": "2025-12-02", "size": 23837068493, "vram": 23837068493, "usage": 406, "io": { "inputs": [ { "nodeId": 58, "nodeType": "LoadImage", "file": "image_z_image_turbo_fun_union_controlnet_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0 }, { "name": "flux_kontext_dev_basic", "title": "Flux Kontext Dev 图像编辑", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "hoverDissolve", "description": "使用Flux Kontext全节点可见性编辑图像，非常适合学习工作流程。", "tutorialUrl": "https://docs.comfy.org/tutorials/flux/flux-1-kontext-dev", "tags": [ "图片编辑" ], "models": [ "Flux.1 Kontext", "Flux" ], "date": "2025-06-26", "size": 17609365914, "vram": 17609365914, "usage": 248, "io": { "inputs": [ { "nodeId": 190, "nodeType": "LoadImage", "file": "flux_kontext_dev_basic_input_image.jpg", "mediaType": "image" } ], "outputs": [ { "nodeId": 136, "nodeType": "SaveImage", "file": "flux_kontext_dev_basic.png", "mediaType": "image" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0 }, { "name": "image_flux2_klein_text_to_image", "title": "Flux.2 [Klein] 4B：文生图", "mediaType": "image", "mediaSubtype": "webp", "description": "BFL 蒸馏模型。亚秒级速度下的卓越画质，非常适合实时生成且保证质量。市场推广将聚焦该模型。", "tags": [ "图像", "文生图" ], "models": [ "Flux", "Flux.2 Klein" ], "date": "2026-01-16", "size": 17072495002, "vram": 9878424781, "usage": 80, "tutorialUrl": "https://docs.comfy.org/tutorials/flux/flux-2-klein", "searchRank": 0, "openSource": true, "io": { "outputs": [ { "nodeId": 78, "nodeType": "SaveImage", "file": "image_flux2_klein_text_to_image.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "output/image_flux2_klein_text_to_image.png" ] }, { "name": "api_reve_image_remix", "title": "Reve：图像混音与编辑", "description": "使用Reve的高级模型重新混合和编辑图像。将多个参考图像合并为单一输出", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图片编辑", "API" ], "models": [ "Reve" ], "logos": [ { "provider": "Reve" } ], "date": "2026-03-14", "openSource": false, "size": 0, "vram": 0, "usage": 47, "searchRank": 0, "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadImage", "file": "iced_drink_cup.png", "mediaType": "image" }, { "nodeId": 4, "nodeType": "LoadImage", "file": "casual_male_portrait.png", "mediaType": "image" }, { "nodeId": 5, "nodeType": "LoadImage", "file": "japanese_street_alley.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 3, "nodeType": "SaveImage", "file": "api_reve_image_remix.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "thumbnail/api_reve_image_remix.png" ] }, { "name": "api_flux2", "title": "Flux.2 Pro", "description": "生成照片级真实感图像，具备多参考一致性与专业文字渲染。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图片编辑", "图像", "API", "文生图" ], "models": [ "Flux.2", "Flux" ], "date": "2025-11-26", "openSource": false, "size": 0, "vram": 0, "usage": 223, "io": { "inputs": [ { "nodeId": 15, "nodeType": "LoadImage", "file": "api_flux2_input_image_1.png", "mediaType": "image" }, { "nodeId": 16, "nodeType": "LoadImage", "file": "api_flux2_input_image_4.png", "mediaType": "image" }, { "nodeId": 18, "nodeType": "LoadImage", "file": "api_flux2_input_image_3.png", "mediaType": "image" }, { "nodeId": 19, "nodeType": "LoadImage", "file": "api_flux2_input_image_2.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "image_flux2_klein_image_edit_4b_distilled", "tutorialUrl": "https://docs.comfy.org/tutorials/flux/flux-2-klein", "title": "Flux.2 [Klein] 4B 蒸馏版：图片编辑", "mediaType": "image", "mediaSubtype": "webp", "description": "Klein 家族中速度最快的版本。专为交互式应用、实时预览和对延迟敏感的生产场景打造。", "tags": [ "图像", "图片编辑" ], "models": [ "Flux", "Flux.2 Klein" ], "date": "2026-01-16", "size": 12455405158, "vram": 12455405158, "usage": 432, "io": { "inputs": [ { "nodeId": 76, "nodeType": "LoadImage", "file": "handbag_white.png", "mediaType": "image" }, { "nodeId": 81, "nodeType": "LoadImage", "file": "comfy_logo_blue.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 94, "nodeType": "SaveImage", "file": "image_flux2_klein_image_edit_4b_distilled.png", "mediaType": "image" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0, "thumbnail": [ "thumbnail/image_flux2_klein_image_edit_4b_distilled.png" ] }, { "name": "flux_fill_inpaint_example", "title": "Flux.1 图像修复", "mediaType": "image", "mediaSubtype": "webp", "description": "使用Flux修复功能填充图像的缺失部分。", "thumbnailVariant": "compareSlider", "tutorialUrl": "https://docs.comfy.org/tutorials/flux/flux-1-fill-dev", "tags": [ "图像", "图像修复" ], "models": [ "Flux", "Flux.1" ], "date": "2025-03-01", "size": 34144990003, "vram": 34144990003, "usage": 323, "io": { "inputs": [ { "nodeId": 17, "nodeType": "LoadImage", "file": "flux_fill_inpaint_example_input_image.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 9, "nodeType": "SaveImage", "file": "flux_fill_inpaint_example.png", "mediaType": "image" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0, "thumbnail": [ "input/flux_fill_inpaint_example_input_image.png", "output/flux_fill_inpaint_example.png" ] }, { "name": "flux1_dev_uso_reference_image_gen", "title": "Flux.1 Dev USO参考图像生成", "description": "使用参考图像来控制风格和主题 - 在改变艺术风格的同时保持角色的脸部，或将艺术风格应用于新场景", "thumbnailVariant": "hoverDissolve", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图像" ], "models": [ "Flux", "Flux.1" ], "date": "2025-09-02", "tutorialUrl": "https://docs.comfy.org/tutorials/flux/flux-1-uso", "size": 18575733555, "vram": 19864223744, "usage": 239, "io": { "inputs": [ { "nodeId": 47, "nodeType": "LoadImage", "file": "flux1_dev_uso_reference_image_gen_input_image.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 9, "nodeType": "SaveImage", "file": "flux1_dev_uso_reference_image_gen.png", "mediaType": "image" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0 }, { "name": "api_reve_image_edit", "title": "Reve：图像编辑", "description": "上传一张图片并提供文字指令进行编辑。使用Reve的高级编辑模型生成修改后的输出图像，用于对象移除、修复或重构等任务。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "图片编辑" ], "models": [ "Reve" ], "logos": [ { "provider": "Reve" } ], "date": "2026-03-14", "openSource": false, "size": 0, "vram": 0, "usage": 24, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 4, "nodeType": "LoadImage", "file": "woman_running_in_field.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 5, "nodeType": "SaveImage", "file": "api_reve_image_edit.png", "mediaType": "image" } ] }, "thumbnail": [ "input/woman_running_in_field.png", "output/api_reve_image_edit.png" ] }, { "name": "Image_capybara_v0_1_text_to_image", "title": "Capybara：文生图", "description": "输入文本提示以生成图像。此工作流程使用Capybara统一模型进行文本到图像合成。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生图", "图像" ], "models": [ "Capybara" ], "date": "2026-02-23", "openSource": true, "size": 36185099469, "vram": 36185099469, "usage": 43, "searchRank": 0, "io": { "outputs": [ { "nodeId": 102, "nodeType": "SaveImage", "file": "Image_capybara_v0_1_text_to_image.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "output/Image_capybara_v0_1_text_to_image.png" ] }, { "name": "image_qwen_image_edit", "title": "Qwen图像编辑", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "description": "使用Qwen-Image-Edit的20B MMDiT模型进行精确双语文本编辑和双重语义/外观编辑功能来编辑图像。", "tutorialUrl": "https://docs.comfy.org/tutorials/image/qwen/qwen-image-edit", "tags": [ "图片编辑" ], "models": [ "Qwen-Image-Edit" ], "date": "2025-08-18", "size": 31782757990, "vram": 31782757990, "usage": 221, "io": { "inputs": [ { "nodeId": 78, "nodeType": "LoadImage", "file": "image_qwen_image_edit_input_image.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 60, "nodeType": "SaveImage", "file": "image_qwen_image_edit.jpg", "mediaType": "image" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0, "thumbnail": [ "input/image_qwen_image_edit_input_image.png", "output/image_qwen_image_edit.jpg" ] }, { "name": "Image_capybara_v0_1_image_edit", "title": "Capybara：图像编辑", "description": "上传图像或视频并输入文本指令。使用Capybara模型生成编辑后的输出，适用于风格更改、对象替换或时间调整等任务。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "图片编辑" ], "models": [ "Capybara" ], "date": "2026-02-23", "openSource": true, "size": 37044092928, "vram": 37044092928, "usage": 99, "searchRank": 0, "io": { "inputs": [ { "nodeId": 80, "nodeType": "LoadImage", "file": "wind_woman.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 102, "nodeType": "SaveImage", "file": "Image_capybara_v0_1_image_edit.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "input/wind_woman.png", "output/Image_capybara_v0_1_image_edit.png" ] }, { "name": "image_longcat_image_edit", "title": "长猫图片编辑", "description": "使用LongCat-Image-Edit模型编辑图像。支持双语指令，用于全局、局部、文本和参考引导编辑，同时保持视觉一致性。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "图像", "图片编辑" ], "models": [ "LongCat Image Edit" ], "date": "2026-03-23", "openSource": true, "size": 22226455757, "vram": 22226455757, "usage": 4, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 13, "nodeType": "LoadImage", "file": "steel_in_the_mist.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 9, "nodeType": "SaveImage", "file": "image_longcat_image_edit.png", "mediaType": "image" } ] }, "thumbnail": [ "input/steel_in_the_mist.png", "output/image_longcat_image_edit.png" ], "includeOnDistributions": [ "local" ] }, { "name": "api_reve_image_create", "title": "Reve：文生图", "description": "输入文本提示以生成具有精细美学和准确文本渲染的高质量图像。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "文生图" ], "models": [ "Reve" ], "logos": [ { "provider": "Reve" } ], "date": "2026-03-13", "openSource": false, "size": 0, "vram": 0, "usage": 19, "searchRank": 0, "io": { "outputs": [ { "nodeId": 2, "nodeType": "SaveImage", "file": "api_reve_image_create.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "output/api_reve_image_create.png" ] }, { "name": "api_openai_dall_e_2_inpaint", "title": "OpenAI：Dall-E 2修复", "description": "使用OpenAI Dall-E 2 API通过修复功能编辑图像。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "图像修复", "图像", "API" ], "models": [ "OpenAI" ], "date": "2025-03-01", "tutorialUrl": "https://docs.comfy.org/tutorials/partner-nodes/openai/dall-e-2", "openSource": false, "size": 0, "vram": 0, "usage": 9, "io": { "inputs": [ { "nodeId": 19, "nodeType": "LoadImage", "file": "api_openai_dall_e_2_inpaint_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_quiver_text_to_svg", "title": "Quiver：文本转SVG", "description": "输入文本提示和可选参考图像，生成一个或多个SVG矢量图形。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "SVG", "矢量" ], "models": [ "Quiver" ], "logos": [ { "provider": "Quiver" } ], "date": "2026-03-23", "openSource": false, "size": 0, "vram": 0, "usage": 14, "searchRank": 0, "username": "ComfyUI", "thumbnail": [ "thumbnail/api_quiver_text_to_svg.png" ] }, { "name": "api_quiver_image_to_svg", "title": "Quiver：图像转SVG", "description": "将图像输入转换为SVG输出。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "SVG", "矢量化" ], "models": [ "Quiver" ], "logos": [ { "provider": "Quiver" } ], "date": "2026-03-23", "openSource": false, "size": 0, "vram": 0, "usage": 25, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadImage", "file": "abstract_horse.png", "mediaType": "image" } ] }, "thumbnail": [ "thumbnail/api_quiver_image_to_svg.png" ] }, { "name": "api_kling_omni_image", "title": "Kling : O1 图像", "description": "使用 Kling O1 从文本提示生成高质量图像，支持最多10张参考图像来指导风格和内容。具有灵活的宽高比和1K/2K分辨率选项。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "角色参考", "文生图" ], "models": [ "Kling O1" ], "date": "2026-01-05", "openSource": false, "usage": 59, "io": { "outputs": [ { "nodeId": 5, "nodeType": "SaveImage", "file": "api_kling_omni_image.png", "mediaType": "image" } ] }, "size": 0, "vram": 0, "searchRank": 0, "username": "ComfyUI", "thumbnail": [ "output/api_kling_omni_image.png" ] }, { "name": "image_lens_t2i", "tutorialUrl": "https://docs.comfy.org/tutorials/image/lens/lens", "title": "镜头：文字转图像", "description": "输入文本提示并选择分辨率和宽高比。使用高效的Lens文本到图像模型生成高质量图像。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生图" ], "models": [ "Lens" ], "date": "2026-05-26", "openSource": true, "size": 21796959027, "vram": 21796959027, "usage": 0, "searchRank": 0, "username": "ComfyUI", "io": { "outputs": [ { "nodeId": 43, "nodeType": "SaveImage", "file": "image_lens_t2i.png", "mediaType": "image" } ] }, "thumbnail": [ "output/image_lens_t2i.png" ] }, { "name": "image_chroma_text_to_image", "title": "Chroma:文生图", "mediaType": "image", "mediaSubtype": "webp", "description": "Chroma是从flux修改而来，在架构上有一些变化。", "tags": [ "文生图", "图像" ], "models": [ "Chroma", "Flux" ], "date": "2025-06-04", "size": 23300197581, "vram": 15569256448, "usage": 224, "io": { "outputs": [ { "nodeId": 740, "nodeType": "SaveImage", "file": "image_chroma_text_to_image.png", "mediaType": "image" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0, "thumbnail": [ "output/image_chroma_text_to_image.png" ] }, { "name": "flux1_krea_dev", "title": "Flux.1 Krea Dev", "mediaType": "image", "mediaSubtype": "webp", "description": "经过微调的FLUX模型，将照片级真实感推向极致", "tutorialUrl": "https://docs.comfy.org/tutorials/flux/flux1-krea-dev", "tags": [ "图像", "文生图" ], "models": [ "Flux", "Flux.1 Krea Dev" ], "date": "2025-07-31", "size": 22226455757, "vram": 22226455757, "usage": 199, "io": { "outputs": [ { "nodeId": 9, "nodeType": "SaveImage", "file": "flux1_krea_dev.png", "mediaType": "image" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0, "thumbnail": [ "output/flux1_krea_dev.png" ] }, { "name": "flux_dev_checkpoint_example", "title": "Flux.1 Dev fp8：文生图", "mediaType": "image", "mediaSubtype": "webp", "description": "使用Flux Dev fp8量化版本生成图像。适用于显存有限的设备，只需要一个模型文件，但图像质量略低于完整版本。", "tutorialUrl": "https://docs.comfy.org/tutorials/flux/flux-1-text-to-image", "tags": [ "文生图", "图像" ], "models": [ "Flux.1", "Flux" ], "date": "2025-03-01", "size": 34144990003, "vram": 34144990003, "usage": 186, "io": { "outputs": [ { "nodeId": 9, "nodeType": "SaveImage", "file": "flux_dev_checkpoint_example.png", "mediaType": "image" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0 }, { "name": "image_flux2_fp8", "tutorialUrl": "https://docs.comfy.org/tutorials/flux/flux-2-dev", "title": "产品样机（Flux.2 Dev FP8）", "mediaType": "image", "mediaSubtype": "webp", "description": "通过使用多参考一致性将设计图案应用到包装、马克杯等产品上，创建产品模型图。", "tags": [ "文生图", "图像", "图片编辑", "样机", "产品" ], "models": [ "Flux.2 Dev", "Flux" ], "date": "2025-11-26", "size": 56586194125, "vram": 56586194125, "usage": 149, "io": { "inputs": [ { "nodeId": 42, "nodeType": "LoadImage", "file": "image_flux2_input_ref_image.png", "mediaType": "image" }, { "nodeId": 46, "nodeType": "LoadImage", "file": "image_flux2_input_Illustration.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 9, "nodeType": "SaveImage", "file": "image_flux2_fp8.png", "mediaType": "image" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0 }, { "name": "sdxl_simple_example", "title": "SDXL简单版", "mediaType": "image", "mediaSubtype": "webp", "description": "使用SDXL生成高质量图像。", "tutorialUrl": "https://comfyanonymous.github.io/ComfyUI_examples/sdxl/", "tags": [ "文生图", "图像" ], "models": [ "SDXL", "Stability" ], "date": "2025-03-01", "size": 13013750907, "vram": 13013750907, "usage": 283, "username": "ComfyUI" }, { "name": "flux_fill_outpaint_example", "title": "Flux.1 外绘", "mediaType": "image", "mediaSubtype": "webp", "description": "使用Flux外绘功能将图像扩展到边界之外。", "thumbnailVariant": "compareSlider", "tutorialUrl": "https://docs.comfy.org/tutorials/flux/flux-1-fill-dev", "tags": [ "外绘", "图像" ], "models": [ "Flux.1", "Flux" ], "date": "2025-03-01", "size": 10372346020, "vram": 10372346020, "usage": 97, "io": { "inputs": [ { "nodeId": 17, "nodeType": "LoadImage", "file": "flux_fill_outpaint_example_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "thumbnail/flux_fill_outpaint_example-1.png", "thumbnail/flux_fill_outpaint_example-2.png" ] }, { "name": "flux_dev_full_text_to_image", "title": "Flux.1 Dev: 文生图", "mediaType": "image", "mediaSubtype": "webp", "description": "使用Flux Dev完整版本生成高质量图像。需要更大的显存和多个模型文件，但提供最佳的提示跟随能力和图像质量。", "tutorialUrl": "https://docs.comfy.org/tutorials/flux/flux-1-text-to-image", "tags": [ "图像", "文生图" ], "models": [ "Flux", "Flux.1" ], "date": "2025-03-01", "size": 34144990003, "vram": 23622320128, "usage": 119, "io": { "outputs": [ { "nodeId": 9, "nodeType": "SaveImage", "file": "flux_dev_full_text_to_image.png", "mediaType": "image" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0, "thumbnail": [ "output/flux_dev_full_text_to_image.png" ] }, { "name": "image_qwen_image_instantx_controlnet", "title": "Qwen-Image InstantX ControlNet", "mediaType": "image", "mediaSubtype": "webp", "description": "使用 Qwen-Image InstantX ControlNet 生成图像，支持 canny、柔和边缘、深度、姿态控制", "tags": [ "图像", "ControlNet" ], "tutorialUrl": "https://docs.comfy.org/tutorials/image/qwen/qwen-image", "models": [ "Qwen-Image" ], "date": "2025-08-23", "size": 35304631173, "vram": 35304631173, "usage": 63, "io": { "inputs": [ { "nodeId": 71, "nodeType": "LoadImage", "file": "image_qwen_image_instantx_controlnet_input_image.jpg", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "image_qwen_image", "title": "Qwen-Image：文本生成图像", "mediaType": "image", "mediaSubtype": "webp", "description": "使用Qwen-Image的20B MMDiT模型生成具有卓越多语言文本渲染和编辑功能的图像。", "tutorialUrl": "https://docs.comfy.org/tutorials/image/qwen/qwen-image", "tags": [ "图像", "文生图" ], "models": [ "Qwen-Image" ], "date": "2025-08-05", "size": 31782757990, "vram": 31782757990, "usage": 49, "io": { "outputs": [ { "nodeId": 60, "nodeType": "SaveImage", "file": "image_qwen_image.png", "mediaType": "image" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0, "thumbnail": [ "output/image_qwen_image.png" ] }, { "name": "image_kandinsky5_t2i", "title": "Kandinsky 5.0 图像 Lite 文本到图像", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图像", "文生图" ], "models": [ "Kandinsky" ], "thumbnailVariant": "hoverZoom", "description": "轻量级 2B 模型，从英语和俄语提示生成高质量视觉图像。", "date": "2026-02-03", "usage": 29, "size": 21689584845, "vram": 21689584845, "searchRank": 0, "openSource": true, "io": { "outputs": [ { "nodeId": 16, "nodeType": "SaveImage", "file": "kandinsky5-t2i.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "output/kandinsky5-t2i.png" ] }, { "name": "image_anima_preview", "tutorialUrl": "https://docs.comfy.org/tutorials/image/anima/anima", "title": "Anima 动漫文生图", "description": "输入文本提示，使用Anima模型生成动漫风格图像。配置步数和CFG比例等设置以控制输出。该工作流程将根据您的描述生成非写实风格的插图。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "动漫", "文生图" ], "models": [ "Anima" ], "date": "2025-05-07", "openSource": true, "usage": 463, "size": 5583457485, "vram": 5583457485, "searchRank": 0, "io": { "outputs": [ { "nodeId": 46, "nodeType": "SaveImage", "file": "image_anima_preview.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "output/image_anima_preview.png" ] }, { "name": "image_chroma1_radiance_text_to_image", "title": "Chroma1 Radiance文生图", "mediaType": "image", "mediaSubtype": "webp", "description": "Chroma1-Radiance 直接处理图像像素而非压缩潜变量，提供更高质量的图像，减少伪影和失真。", "tags": [ "图像", "文生图" ], "models": [ "Chroma" ], "date": "2025-09-18", "size": 24159191040, "vram": 24159191040, "usage": 66, "io": { "outputs": [ { "nodeId": 740, "nodeType": "SaveImage", "file": "image_chroma1_radiance_text_to_image.png", "mediaType": "image" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0, "thumbnail": [ "output/image_chroma1_radiance_text_to_image.png" ] }, { "name": "image_qwen_image_edit_2509_relight", "tutorialUrl": "https://docs.comfy.org/tutorials/image/qwen/qwen-image-edit", "title": "照片重打光", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "description": "使用 Qwen-Image-Edit 和 LoRA 重新打光图片。", "tags": [ "图片编辑", "重打光" ], "models": [ "Qwen-Image-Edit" ], "date": "2025-12-15", "size": 31772020572, "vram": 31772020572, "usage": 112, "io": { "inputs": [ { "nodeId": 11, "nodeType": "LoadImage", "file": "warm_living_room.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "image_qwen_image_instantx_inpainting_controlnet", "title": "Qwen-Image InstantX 修复 ControlNet", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "description": "使用 Qwen-Image InstantX ControlNet 进行专业图像修复和编辑。支持对象替换、文本修改、背景更改和外扩绘制。", "tags": [ "图像", "ControlNet", "图像修复" ], "tutorialUrl": "https://docs.comfy.org/tutorials/image/qwen/qwen-image", "models": [ "Qwen-Image" ], "date": "2025-09-12", "size": 36013300777, "vram": 36013300777, "usage": 146, "io": { "inputs": [ { "nodeId": 71, "nodeType": "LoadImage", "file": "image_qwen_image_instantx_inpainting_controlnet_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_from_photo_2_miniature", "title": "照片转模型风格", "description": "将真实建筑照片转化为建筑蓝图，再转化为精致的实体比例模型。实现从照片到微缩模型的完整建筑可视化流程。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "图片编辑", "图像", "3D" ], "models": [ "Gemini3 Pro Image Preview", "Nano Banana Pro", "Google" ], "date": "2025-11-21", "openSource": false, "size": 0, "vram": 0, "usage": 87, "io": { "inputs": [ { "nodeId": 11, "nodeType": "LoadImage", "file": "api_from_photo_2_miniature_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_bria_image_edit", "title": "Bria：图像编辑", "description": "FIBO通过结构化的JSON控制照明、构图和相机设置，实现精确、可预测的图像编辑", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "图片编辑" ], "models": [ "BRIA" ], "date": "2026-01-20", "openSource": false, "usage": 34, "io": { "inputs": [ { "nodeId": 5, "nodeType": "LoadImage", "file": "sunlit_rooftop_capture.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 8, "nodeType": "SaveImage", "file": "api_bria_image_edit.png", "mediaType": "image" } ] }, "username": "ComfyUI", "logos": [ { "provider": "BRIA" } ], "size": 0, "vram": 0, "searchRank": 0, "thumbnail": [ "input/sunlit_rooftop_capture.png", "output/api_bria_image_edit.png" ] }, { "name": "api_bria_image_outpainting", "title": "Bria：图像外绘", "description": "一款专业的Bria驱动外绘模板，专为无缝图像扩展和边缘感知场景扩展而设计", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "外绘" ], "models": [ "BRIA" ], "date": "2026-01-20", "openSource": false, "usage": 18, "io": { "inputs": [ { "nodeId": 11, "nodeType": "LoadImage", "file": "baby_otter.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "thumbnail/api_bria_image_outpainting-1.png", "thumbnail/api_bria_image_outpainting-2.png" ], "logos": [ { "provider": "BRIA" } ], "size": 0, "vram": 0, "searchRank": 0 }, { "name": "sd3.5_large_canny_controlnet_example", "title": "SD3.5大型Canny ControlNet", "mediaType": "image", "mediaSubtype": "webp", "description": "使用SD 3.5 Canny ControlNet边缘检测引导生成图像。", "thumbnailVariant": "hoverDissolve", "tutorialUrl": "https://comfyanonymous.github.io/ComfyUI_examples/sd3/#sd35-controlnets", "tags": [ "图像", "ControlNet" ], "models": [ "SD3.5", "Stability" ], "date": "2025-03-01", "size": 23590107873, "vram": 23590107873, "usage": 28, "io": { "inputs": [ { "nodeId": 45, "nodeType": "LoadImage", "file": "sd3.5_large_canny_controlnet_example_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "image_chrono_edit_14B", "title": "ChronoEdit 14B：图像编辑", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "description": "基于视频模型对图像的动态理解对图像进行编辑，创造符合物理规律的新图像并保持角色和风格的一致性。", "tags": [ "图片编辑" ], "models": [ "Wan2.1", "ChronoEdit", "Nvidia" ], "date": "2025-11-03", "size": 41446434406, "vram": 41446434406, "usage": 45, "io": { "inputs": [ { "nodeId": 52, "nodeType": "LoadImage", "file": "image_chrono_edit_input_image.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 60, "nodeType": "SaveImage", "file": "image_chrono_edit_14B.png", "mediaType": "image" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0, "thumbnail": [ "input/image_chrono_edit_input_image.png", "output/image_chrono_edit_14B.png" ] }, { "name": "image_flux.1_fill_dev_OneReward", "tutorialUrl": "https://docs.comfy.org/tutorials/flux/flux-1-fill-dev", "title": "Flux.1 Dev OneReward", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "description": "支持图像修复、外绘和对象移除等多种任务，由 bytedance-research 团队开发", "tags": [ "图像修复", "外绘" ], "models": [ "Flux.1", "Flux" ], "date": "2025-09-21", "size": 22333829939, "vram": 22333829939, "usage": 62, "io": { "inputs": [ { "nodeId": 17, "nodeType": "LoadImage", "file": "image_flux.1_fill_dev_OneReward_input_image.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 62, "nodeType": "SaveImage", "file": "image_flux.1_fill_dev_OneReward.png", "mediaType": "image" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0 }, { "name": "sd3.5_simple_example", "title": "SD3.5简单版", "mediaType": "image", "mediaSubtype": "webp", "description": "使用SD 3.5生成图像。", "tutorialUrl": "https://comfyanonymous.github.io/ComfyUI_examples/sd3/#sd35", "tags": [ "文生图", "图像" ], "models": [ "SD3.5", "Stability" ], "date": "2025-03-01", "size": 14935748772, "vram": 14935748772, "usage": 73, "username": "ComfyUI" }, { "name": "api_wan_text_to_image", "title": "Wan2.5: 文生图", "description": "使用 FLUX.1 Pro 生成具有出色提示跟随和视觉质量的图像。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生图", "图像", "API" ], "models": [ "Wan2.5", "Wan" ], "date": "2025-09-25", "openSource": false, "size": 0, "vram": 0, "usage": 69, "username": "ComfyUI" }, { "name": "flux_canny_model_example", "title": "Flux.1 Canny 模型", "mediaType": "image", "mediaSubtype": "webp", "description": "使用Flux Canny边缘检测引导生成图像。", "thumbnailVariant": "hoverDissolve", "tutorialUrl": "https://docs.comfy.org/tutorials/flux/flux-1-controlnet", "tags": [ "ControlNet", "图像" ], "models": [ "Flux.1", "Flux" ], "date": "2025-03-01", "size": 34177202258, "vram": 34177202258, "usage": 125, "io": { "inputs": [ { "nodeId": 17, "nodeType": "LoadImage", "file": "flux_canny_model_example_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "image_netayume_lumina_t2i", "title": "NetaYume Lumina 文生图", "mediaType": "image", "mediaSubtype": "webp", "description": "高质量动漫风格图像生成，具有增强的角色理解和细节纹理。基于 Danbooru 数据集从 Neta Lumina 微调。", "tags": [ "动漫", "图像", "文生图" ], "models": [ "OmniGen" ], "date": "2025-10-10", "size": 10630044058, "vram": 10630044058, "usage": 96, "tutorialUrl": "https://docs.comfy.org/tutorials/image/newbie-image/newbie-image-exp-0-1", "searchRank": 0, "openSource": true, "io": { "outputs": [ { "nodeId": 9, "nodeType": "SaveImage", "file": "image_netayume_lumina_t2i.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "output/image_netayume_lumina_t2i.png" ] }, { "name": "api_runway_reference_to_image", "title": "Runway：参考到图像", "description": "使用Runway的AI基于参考风格和构图生成新图像。", "mediaType": "image", "thumbnailVariant": "compareSlider", "mediaSubtype": "webp", "tags": [ "图像", "API" ], "models": [ "Runway" ], "date": "2025-03-01", "openSource": false, "size": 0, "vram": 0, "usage": 11, "io": { "inputs": [ { "nodeId": 5, "nodeType": "LoadImage", "file": "api_runway_reference_to_image_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_stability_ai_i2i", "title": "Stability AI：图像到图像", "description": "使用Stability AI进行高质量生成转换图像，非常适合专业编辑和风格迁移。", "mediaType": "image", "thumbnailVariant": "compareSlider", "mediaSubtype": "webp", "tags": [ "图像", "API" ], "models": [ "Stability" ], "date": "2025-03-01", "openSource": false, "size": 0, "vram": 0, "usage": 22, "io": { "inputs": [ { "nodeId": 5, "nodeType": "LoadImage", "file": "api_stability_ai_i2i_input_image.jpg", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "flux_depth_lora_example", "title": "Flux.1 深度 Lora", "mediaType": "image", "mediaSubtype": "webp", "description": "使用Flux LoRA深度信息引导生成图像。", "thumbnailVariant": "compareSlider", "tutorialUrl": "https://docs.comfy.org/tutorials/flux/flux-1-controlnet", "tags": [ "ControlNet", "图像" ], "models": [ "Flux", "Flux.1" ], "date": "2025-03-01", "size": 35433480192, "vram": 35433480192, "usage": 30, "io": { "inputs": [ { "nodeId": 17, "nodeType": "LoadImage", "file": "flux_depth_lora_example_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0 }, { "name": "api_openai_fashion_billboard_generator", "title": "时尚广告牌生成器", "description": "将服装照片转为专业时尚模特的商场广告。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图片编辑", "图像", "API", "时尚", "样机" ], "models": [ "GPT-Image-1.5", "OpenAI" ], "date": "2025-12-18", "openSource": false, "size": 0, "vram": 0, "usage": 56, "io": { "inputs": [ { "nodeId": 3, "nodeType": "LoadImage", "file": "outfit-streetwear.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "thumbnail/api_openai_fashion_billboard_generator.png" ] }, { "name": "sdxlturbo_example", "title": "SDXL Turbo", "mediaType": "image", "mediaSubtype": "webp", "description": "使用SDXL Turbo单步生成图像。", "tutorialUrl": "https://comfyanonymous.github.io/ComfyUI_examples/sdturbo/", "tags": [ "文生图", "图像" ], "models": [ "SDXL", "Stability" ], "date": "2025-03-01", "size": 6936372183, "vram": 6936372183, "usage": 133, "username": "ComfyUI" }, { "name": "image_qwen_image_union_control_lora", "title": "Qwen-Image Union Control", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "description": "使用Qwen-Image统一的ControlNet LoRA进行精确结构控制生成图像。支持多种控制类型，包括canny、depth、lineart、softedge、normal和openpose，适用于各种创意应用。", "tags": [ "文生图", "图像", "ControlNet" ], "tutorialUrl": "https://docs.comfy.org/tutorials/image/qwen/qwen-image", "models": [ "Qwen-Image" ], "date": "2025-08-23", "size": 32749125632, "vram": 32749125632, "usage": 19, "io": { "inputs": [ { "nodeId": 73, "nodeType": "LoadImage", "file": "image_qwen_image_union_control_lora_input_image.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 60, "nodeType": "SaveImage", "file": "image_qwen_image_union_control_lora.png", "mediaType": "image" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0 }, { "name": "image_qwen_image_controlnet_patch", "title": "Qwen-Image ControlNet 模型补丁", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "description": "使用 Qwen-Image ControlNet 模型控制图像生成。通过模型修补支持 canny、深度和修复控制。", "tutorialUrl": "https://docs.comfy.org/tutorials/image/qwen/qwen-image", "tags": [ "文生图", "图像", "ControlNet" ], "models": [ "Qwen-Image" ], "date": "2025-08-24", "size": 34037615821, "vram": 34037615821, "usage": 45, "io": { "inputs": [ { "nodeId": 71, "nodeType": "LoadImage", "file": "image_qwen_image_controlnet_patch_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_bfl_flux_1_kontext_pro_image", "title": "Flux.1 Kontext Pro", "description": "使用Flux.1 Kontext专业版图像编辑图像。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tutorialUrl": "https://docs.comfy.org/tutorials/partner-nodes/black-forest-labs/flux-1-kontext", "tags": [ "API", "图像", "图片编辑" ], "models": [ "Flux", "Flux.1 Kontext" ], "date": "2025-05-29", "openSource": false, "size": 0, "vram": 0, "usage": 81, "io": { "inputs": [ { "nodeId": 84, "nodeType": "LoadImage", "file": "api_bfl_flux_1_kontext_pro_image_input_image.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 85, "nodeType": "SaveImage", "file": "api_bfl_flux_1_kontext_pro_image.png", "mediaType": "image" } ] }, "username": "ComfyUI", "logos": [ { "provider": "Black Forest Labs" } ], "searchRank": 0, "thumbnail": [ "input/api_bfl_flux_1_kontext_pro_image_input_image.png", "output/api_bfl_flux_1_kontext_pro_image.png" ] }, { "name": "image_newbieimage_exp0_1-t2i", "tutorialUrl": "https://docs.comfy.org/tutorials/image/newbie-image/newbie-image-exp-0-1", "title": "NewBie Exp0.1：动漫生成", "mediaType": "image", "mediaSubtype": "webp", "description": "使用NewBie Exp0.1生成精细的动漫风格图片，支持XML结构化提示词以增强多角色场景与属性控制。", "tags": [ "动漫", "图像", "文生图" ], "models": [ "NewBie" ], "date": "2025-12-19", "size": 16213501542, "vram": 16213501542, "usage": 15, "openSource": true, "searchRank": 0, "io": { "outputs": [ { "nodeId": 9, "nodeType": "SaveImage", "file": "image_newbieimage_exp0_1-t2i.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "output/image_newbieimage_exp0_1-t2i.png" ] }, { "name": "api_runway_text_to_image", "title": "Runway：文生图", "description": "使用Runway的AI模型从文本提示生成高质量图像。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生图", "图像", "API" ], "models": [ "Runway" ], "date": "2025-03-01", "openSource": false, "size": 0, "vram": 0, "usage": 10, "io": { "inputs": [ { "nodeId": 5, "nodeType": "LoadImage", "file": "example.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_luma_photon_i2i", "title": "Luma Photon：图像到图像", "description": "使用图像和提示的组合引导图像生成。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "图像", "API" ], "models": [ "Luma" ], "date": "2025-03-01", "openSource": false, "size": 0, "vram": 0, "usage": 17, "io": { "inputs": [ { "nodeId": 3, "nodeType": "LoadImage", "file": "api_luma_photon_i2i_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_magnific_image_style_transfer", "title": "Magnific：风格参考", "description": "上传目标图像和风格参考图像。生成一张新图像，将参考图像的艺术风格与目标图像的结构融合。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "风格参考" ], "models": [ "Magnific" ], "date": "2026-01-26", "openSource": false, "usage": 61, "io": { "inputs": [ { "nodeId": 11, "nodeType": "LoadImage", "file": "girl_riding_unicorn.png", "mediaType": "image" }, { "nodeId": 13, "nodeType": "LoadImage", "file": "magnific_style_reference.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 14, "nodeType": "SaveImage", "file": "api_magnific_image_style_transfer.png", "mediaType": "image" } ] }, "username": "ComfyUI", "size": 0, "vram": 0, "searchRank": 0, "logos": [ { "provider": "Magnific" } ], "thumbnail": [ "input/girl_riding_unicorn.png", "output/api_magnific_image_style_transfer.png" ] }, { "name": "api_stability_ai_sd3.5_i2i", "title": "Stability AI：SD3.5图像到图像", "description": "生成具有出色提示遵循性的高质量图像。适用于1百万像素分辨率的专业用例。", "mediaType": "image", "thumbnailVariant": "compareSlider", "mediaSubtype": "webp", "tags": [ "图像", "API" ], "models": [ "Stability" ], "date": "2025-03-01", "openSource": false, "size": 0, "vram": 0, "usage": 4, "io": { "inputs": [ { "nodeId": 11, "nodeType": "LoadImage", "file": "api_stability_ai_sd3.5_i2i_input_image.jpg", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "sd3.5_large_depth", "title": "SD3.5大型深度", "mediaType": "image", "mediaSubtype": "webp", "description": "使用SD 3.5深度信息引导生成图像。", "thumbnailVariant": "hoverDissolve", "tutorialUrl": "https://comfyanonymous.github.io/ComfyUI_examples/sd3/#sd35-controlnets", "tags": [ "图像", "ControlNet" ], "models": [ "SD3.5", "Stability" ], "date": "2025-03-01", "size": 23590107873, "vram": 23590107873, "usage": 15, "io": { "inputs": [ { "nodeId": 45, "nodeType": "LoadImage", "file": "sd3.5_large_depth_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_bfl_flux_1_kontext_multiple_images_input", "title": "Flux.1 Kontext 多图像输入", "description": "输入多个图像并使用Flux.1 Kontext进行编辑。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tutorialUrl": "https://docs.comfy.org/tutorials/partner-nodes/black-forest-labs/flux-1-kontext", "tags": [ "API", "图片编辑", "图像" ], "models": [ "Flux.1 Kontext", "Flux" ], "date": "2025-05-29", "openSource": false, "size": 0, "vram": 0, "usage": 40, "io": { "inputs": [ { "nodeId": 84, "nodeType": "LoadImage", "file": "api_bfl_flux_1_kontext_multiple_images_input_girl.jpg", "mediaType": "image" }, { "nodeId": 91, "nodeType": "LoadImage", "file": "api_bfl_flux_1_kontext_multiple_images_input_sofa.jpg", "mediaType": "image" }, { "nodeId": 96, "nodeType": "LoadImage", "file": "api_bfl_flux_1_kontext_multiple_images_input_dog.jpg", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "flux_redux_model_example", "title": "Flux.1 Redux 模型", "mediaType": "image", "mediaSubtype": "webp", "description": "使用Flux Redux从参考图像转移风格生成图像。", "tutorialUrl": "https://docs.comfy.org/tutorials/flux/flux-1-controlnet", "tags": [ "ControlNet", "图像" ], "models": [ "Flux.1", "Flux" ], "date": "2025-03-01", "size": 35154307318, "vram": 35154307318, "usage": 61, "io": { "inputs": [ { "nodeId": 40, "nodeType": "LoadImage", "file": "flux_redux_model_example_reference_image_1.png", "mediaType": "image" }, { "nodeId": 47, "nodeType": "LoadImage", "file": "flux_redux_model_example_reference_image_2.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "flux_schnell", "title": "Flux.1 Schnell FP8", "mediaType": "image", "mediaSubtype": "webp", "description": "使用Flux Schnell fp8量化版本快速生成图像。适用于低端硬件，仅需4步即可生成图像。", "tutorialUrl": "https://docs.comfy.org/tutorials/flux/flux-1-text-to-image", "tags": [ "文生图", "图像" ], "models": [ "Flux.1", "Flux" ], "date": "2025-03-01", "size": 17233556275, "vram": 18253611008, "usage": 34, "username": "ComfyUI" }, { "name": "flux_schnell_full_text_to_image", "title": "Flux.1 Schnell Full：文生图", "mediaType": "image", "mediaSubtype": "webp", "description": "使用Flux Schnell完整版本快速生成图像。使用Apache2.0许可证，仅需4步即可生成图像，同时保持良好的图像质量。", "tutorialUrl": "https://docs.comfy.org/tutorials/flux/flux-1-text-to-image", "tags": [ "文生图", "图像" ], "models": [ "Flux.1", "Flux" ], "date": "2025-03-01", "size": 34144990003, "vram": 34144990003, "usage": 23, "io": { "outputs": [ { "nodeId": 9, "nodeType": "SaveImage", "file": "flux_schnell_full_text_to_image.png", "mediaType": "image" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0, "thumbnail": [ "output/flux_schnell_full_text_to_image.png" ] }, { "name": "api_bfl_flux_pro_t2i", "title": "Flux.1 [Pro]: 文生图", "description": "使用FLUX.1 Pro生成具有出色提示跟随和视觉质量的图像。", "mediaType": "image", "mediaSubtype": "webp", "tutorialUrl": "https://docs.comfy.org/tutorials/partner-nodes/black-forest-labs/flux-1-1-pro-ultra-image", "tags": [ "API", "图像", "图片编辑" ], "models": [ "Flux", "Flux.1" ], "date": "2025-05-01", "openSource": false, "size": 0, "vram": 0, "usage": 41, "io": { "inputs": [ { "nodeId": 14, "nodeType": "LoadImage", "file": "example.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 12, "nodeType": "SaveImage", "file": "api_bfl_flux_pro_t2i.png", "mediaType": "image" } ] }, "username": "ComfyUI", "searchRank": 0, "logos": [ { "provider": "Black Forest Labs" } ], "thumbnail": [ "output/api_bfl_flux_pro_t2i.png" ] }, { "name": "image_pixeldit_t2i", "title": "PixelDiT：文本到图像", "description": "输入文本提示和可选的负面提示。使用PixelDiT的无VAE像素扩散变换器生成1024px图像。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生图" ], "models": [ "PixelDiT" ], "date": "2026-04-29", "openSource": true, "size": 7838315315, "vram": 7838315315, "usage": 0, "searchRank": 0, "username": "ComfyUI", "io": { "outputs": [ { "nodeId": 43, "nodeType": "SaveImage", "file": "image_pixeldit_t2i.png", "mediaType": "image" } ] } }, { "name": "hidream_e1_full", "title": "HiDream E1图像编辑", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "description": "使用HiDream E1编辑图像 - 专业的自然语言图像编辑模型。", "tutorialUrl": "https://docs.comfy.org/tutorials/image/hidream/hidream-e1", "tags": [ "图片编辑", "图像" ], "models": [ "HiDream" ], "date": "2025-05-01", "size": 34252364186, "vram": 34252364186, "usage": 23, "io": { "inputs": [ { "nodeId": 13, "nodeType": "LoadImage", "file": "hidream_e1_full_input_image.jpg", "mediaType": "image" } ], "outputs": [ { "nodeId": 9, "nodeType": "SaveImage", "file": "hidream_e1_full.png", "mediaType": "image" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0, "thumbnail": [ "input/hidream_e1_full_input_image.jpg", "output/hidream_e1_full.png" ] }, { "name": "hidream_e1_1", "title": "HiDream E1.1图像编辑", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "description": "使用HiDream E1.1编辑图像 - 在图像质量和编辑准确性方面优于HiDream-E1-Full。", "tutorialUrl": "https://docs.comfy.org/tutorials/image/hidream/hidream-e1", "tags": [ "图像", "图片编辑" ], "models": [ "HiDream" ], "date": "2025-07-21", "size": 50465865728, "vram": 50465865728, "usage": 42, "io": { "inputs": [ { "nodeId": 13, "nodeType": "LoadImage", "file": "hidream_e1_1_input_image.jpg", "mediaType": "image" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0 }, { "name": "hidream_i1_dev", "title": "HiDream I1开发版", "mediaType": "image", "mediaSubtype": "webp", "description": "使用HiDream I1开发版生成图像 - 平衡版本，具有28个推理步骤，适合中端硬件。", "tutorialUrl": "https://docs.comfy.org/tutorials/image/hidream/hidream-i1", "tags": [ "文生图", "图像" ], "models": [ "HiDream" ], "date": "2025-04-17", "size": 33285996544, "vram": 33285996544, "usage": 11, "io": { "outputs": [ { "nodeId": 9, "nodeType": "SaveImage", "file": "hidream_i1_dev.png", "mediaType": "image" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0, "thumbnail": [ "output/hidream_i1_dev.png" ] }, { "name": "hidream_i1_full", "title": "HiDream I1完整版", "mediaType": "image", "mediaSubtype": "webp", "description": "使用HiDream I1完整版生成图像 - 完整版本，具有50个推理步骤，输出质量最高。", "tutorialUrl": "https://docs.comfy.org/tutorials/image/hidream/hidream-i1", "tags": [ "文生图", "图像" ], "models": [ "HiDream" ], "date": "2025-04-17", "size": 24266565222, "vram": 24266565222, "usage": 20, "io": { "outputs": [ { "nodeId": 90, "nodeType": "SaveImage", "file": "hidream_i1_full.png", "mediaType": "image" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0, "thumbnail": [ "output/hidream_i1_full.png" ] }, { "name": "hidream_i1_fast", "title": "HiDream I1快速版", "mediaType": "image", "mediaSubtype": "webp", "description": "使用HiDream I1快速版快速生成图像 - 轻量级版本，具有16个推理步骤，非常适合在低端硬件上快速预览。", "tutorialUrl": "https://docs.comfy.org/tutorials/image/hidream/hidream-i1", "tags": [ "文生图", "图像" ], "models": [ "HiDream" ], "date": "2025-04-17", "size": 24266565222, "vram": 24266565222, "usage": 2, "io": { "outputs": [ { "nodeId": 88, "nodeType": "SaveImage", "file": "hidream_i1_fast.png", "mediaType": "image" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0, "thumbnail": [ "output/hidream_i1_fast.png" ] }, { "name": "api_stability_ai_stable_image_ultra_t2i", "title": "Stability AI：Stable Image Ultra文生图", "description": "生成具有出色提示遵循性的高质量图像。适用于1百万像素分辨率的专业用例。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "图像", "文生图" ], "models": [ "Stability" ], "date": "2025-03-01", "openSource": false, "size": 0, "vram": 0, "usage": 8, "io": { "inputs": [ { "nodeId": 57, "nodeType": "LoadImage", "file": "example.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 43, "nodeType": "SaveImage", "file": "api_stability_ai_stable_image_ultra_t2i.png", "mediaType": "image" } ] }, "username": "ComfyUI", "logos": [ { "provider": "Stability" } ], "searchRank": 0, "thumbnail": [ "output/api_stability_ai_stable_image_ultra_t2i.png" ] }, { "name": "api_openai_image_1_i2i", "title": "OpenAI：GPT-Image-1图像到图像", "description": "使用OpenAI GPT Image 1 API从输入图像生成图像。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "图像", "API" ], "models": [ "GPT-Image-1", "OpenAI" ], "date": "2025-03-01", "tutorialUrl": "https://docs.comfy.org/tutorials/partner-nodes/openai/gpt-image-1", "openSource": false, "size": 0, "vram": 0, "usage": 22, "io": { "inputs": [ { "nodeId": 14, "nodeType": "LoadImage", "file": "api_openai_image_1_input_image.jpg", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "image_omnigen2_image_edit", "title": "OmniGen2图像编辑", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "description": "使用OmniGen2的高级图像编辑功能和文本渲染支持，通过自然语言指令编辑图像。", "tutorialUrl": "https://docs.comfy.org/tutorials/image/omnigen/omnigen2", "tags": [ "图像", "图片编辑" ], "models": [ "OmniGen" ], "date": "2025-06-30", "size": 15784004813, "vram": 15784004813, "usage": 35, "io": { "inputs": [ { "nodeId": 16, "nodeType": "LoadImage", "file": "image_omnigen2_image_edit_input_image.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 9, "nodeType": "SaveImage", "file": "image_omnigen2_image_edit.png", "mediaType": "image" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0, "thumbnail": [ "input/image_omnigen2_image_edit_input_image.png", "output/image_omnigen2_image_edit.png" ] }, { "name": "sdxl_revision_text_prompts", "title": "SDXL修订文本提示", "mediaType": "image", "mediaSubtype": "webp", "description": "使用SDXL修订版从参考图像转移概念生成图像。", "tutorialUrl": "https://comfyanonymous.github.io/ComfyUI_examples/sdxl/#revision", "tags": [ "文生图", "图像" ], "models": [ "SDXL", "Stability" ], "date": "2025-03-01", "size": 10630044058, "vram": 10630044058, "usage": 55, "io": { "inputs": [ { "nodeId": 34, "nodeType": "LoadImage", "file": "sdxl_revision_text_prompts_ref_image1.png", "mediaType": "image" }, { "nodeId": 38, "nodeType": "LoadImage", "file": "sdxl_revision_text_prompts_ref_image2.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_openai_image_1_inpaint", "title": "OpenAI：GPT-Image-1修复", "description": "使用OpenAI GPT Image 1 API通过修复功能编辑图像。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "图像修复", "图像", "API" ], "models": [ "GPT-Image-1", "OpenAI" ], "date": "2025-03-01", "tutorialUrl": "https://docs.comfy.org/tutorials/partner-nodes/openai/gpt-image-1", "openSource": false, "size": 0, "vram": 0, "usage": 36, "io": { "inputs": [ { "nodeId": 29, "nodeType": "LoadImage", "file": "api_openai_image_1_input_image_mask.jpg", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_openai_image_1_t2i", "title": "OpenAI：GPT-Image-1文生图", "description": "使用OpenAI GPT Image 1 API从文本提示生成图像。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "图像", "文生图" ], "models": [ "GPT-Image-1", "OpenAI" ], "date": "2025-03-01", "tutorialUrl": "https://docs.comfy.org/tutorials/partner-nodes/openai/gpt-image-1", "openSource": false, "size": 0, "vram": 0, "usage": 3, "io": { "inputs": [ { "nodeId": 16, "nodeType": "LoadImage", "file": "example.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 14, "nodeType": "SaveImage", "file": "api_openai_image_1_t2i.png", "mediaType": "image" } ] }, "username": "ComfyUI", "logos": [ { "provider": "OpenAI" } ], "searchRank": 0, "thumbnail": [ "output/api_openai_image_1_t2i.png" ] }, { "name": "api_openai_dall_e_2_t2i", "title": "OpenAI：Dall-E 2文生图", "description": "使用OpenAI Dall-E 2 API从文本提示生成图像。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "图像", "文生图" ], "models": [ "OpenAI" ], "date": "2025-03-01", "tutorialUrl": "https://docs.comfy.org/tutorials/partner-nodes/openai/dall-e-2", "openSource": false, "size": 0, "vram": 0, "usage": 3, "io": { "outputs": [ { "nodeId": 20, "nodeType": "SaveImage", "file": "api_openai_dall_e_2_t2i.png", "mediaType": "image" } ] }, "username": "ComfyUI", "logos": [ { "provider": "OpenAI" } ], "searchRank": 0, "thumbnail": [ "output/api_openai_dall_e_2_t2i.png" ] }, { "name": "api_openai_dall_e_3_t2i", "title": "OpenAI：Dall-E 3文生图", "description": "使用OpenAI Dall-E 3 API从文本提示生成图像。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "图像", "文生图" ], "models": [ "OpenAI" ], "date": "2025-03-01", "tutorialUrl": "https://docs.comfy.org/tutorials/partner-nodes/openai/dall-e-3", "openSource": false, "size": 0, "vram": 0, "usage": 12, "io": { "outputs": [ { "nodeId": 24, "nodeType": "SaveImage", "file": "api_openai_dall_e_3_t2i.png", "mediaType": "image" } ] }, "username": "ComfyUI", "logos": [ { "provider": "OpenAI" } ], "searchRank": 0, "thumbnail": [ "output/api_openai_dall_e_3_t2i.png" ] }, { "name": "image_ovis_text_to_image", "tutorialUrl": "https://docs.comfy.org/tutorials/image/ovis/ovis-image", "title": "Ovis-Image 文生像", "mediaType": "image", "mediaSubtype": "webp", "description": "Ovis-Image 是一个 7B 参数的文本生成图像模型，专门针对生成图像中的高质量文本渲染进行了优化。该模型设计用于在计算资源受限的情况下高效运行，擅长准确生成包含文本内容的图像。", "tags": [ "文生图", "图像" ], "models": [ "Ovis-Image" ], "date": "2025-12-02", "size": 20228222222, "vram": 20228222222, "usage": 21, "username": "ComfyUI" }, { "name": "api_recraft_image_gen_with_color_control", "title": "Recraft：色彩控制图像生成", "description": "使用Recraft生成具有自定义调色板和品牌特定视觉效果的图像。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生图", "图像", "API" ], "models": [ "Recraft" ], "date": "2025-03-01", "openSource": false, "size": 0, "vram": 0, "usage": 5, "username": "ComfyUI" }, { "name": "api_recraft_vector_gen", "title": "Recraft：矢量生成", "description": "使用Recraft的AI矢量生成器从文本提示生成高质量矢量图像。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生图", "图像", "API", "矢量" ], "models": [ "Recraft" ], "date": "2025-03-01", "openSource": false, "size": 0, "vram": 0, "usage": 5, "username": "ComfyUI" }, { "name": "api_recraft_image_gen_with_style_control", "title": "Recraft：风格控制图像生成", "description": "通过视觉示例控制风格，调整定位并微调对象。存储和分享风格以实现完美的品牌一致性。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生图", "图像", "API" ], "models": [ "Recraft" ], "date": "2025-03-01", "openSource": false, "size": 0, "vram": 0, "usage": 12, "username": "ComfyUI" }, { "name": "sdxl_refiner_prompt_example", "title": "SDXL精炼提示", "mediaType": "image", "mediaSubtype": "webp", "description": "使用精炼模型增强SDXL图像。", "tutorialUrl": "https://comfyanonymous.github.io/ComfyUI_examples/sdxl/", "tags": [ "文生图", "图像" ], "models": [ "SDXL", "Stability" ], "date": "2025-03-01", "size": 13013750907, "vram": 13013750907, "usage": 52, "username": "ComfyUI" }, { "name": "api_bfl_flux_1_kontext_max_image", "title": "Flux.1 Kontext Max", "description": "使用Flux.1 Kontext最大版图像编辑图像。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tutorialUrl": "https://docs.comfy.org/tutorials/partner-nodes/black-forest-labs/flux-1-kontext", "tags": [ "API", "图片编辑", "图像" ], "models": [ "Flux.1 Kontext", "Flux" ], "date": "2025-05-29", "openSource": false, "size": 0, "vram": 0, "usage": 72, "io": { "inputs": [ { "nodeId": 84, "nodeType": "LoadImage", "file": "api_bfl_flux_1_kontext_max_image_style_ref.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_openai_image_1_multi_inputs", "title": "OpenAI：GPT-Image-1多输入", "description": "使用OpenAI GPT Image 1 API从多个输入生成图像。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "文生图", "图像", "API" ], "models": [ "GPT-Image-1", "OpenAI" ], "date": "2025-03-01", "tutorialUrl": "https://docs.comfy.org/tutorials/partner-nodes/openai/gpt-image-1", "openSource": false, "size": 0, "vram": 0, "usage": 29, "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadImage", "file": "api_openai_image_1_hat.png", "mediaType": "image" }, { "nodeId": 4, "nodeType": "LoadImage", "file": "api_openai_image_1_input_image.jpg", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "image_omnigen2_t2i", "title": "OmniGen2：文本到图像", "mediaType": "image", "mediaSubtype": "webp", "description": "使用OmniGen2统一的7B多模态模型和双路径架构从文本提示生成高质量图像。", "tutorialUrl": "https://docs.comfy.org/tutorials/image/omnigen/omnigen2", "tags": [ "图像", "文生图" ], "models": [ "OmniGen" ], "date": "2025-06-30", "size": 15784004813, "vram": 15784004813, "usage": 9, "io": { "outputs": [ { "nodeId": 9, "nodeType": "SaveImage", "file": "image_omnigen2_t2i.png", "mediaType": "image" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0, "thumbnail": [ "output/image_omnigen2_t2i.png" ] }, { "name": "sd3.5_large_blur", "title": "SD3.5大型模糊", "mediaType": "image", "mediaSubtype": "webp", "description": "使用SD 3.5模糊参考图像引导生成图像。", "thumbnailVariant": "hoverDissolve", "tutorialUrl": "https://comfyanonymous.github.io/ComfyUI_examples/sd3/#sd35-controlnets", "tags": [ "图像" ], "models": [ "SD3.5", "Stability" ], "date": "2025-03-01", "size": 23590107873, "vram": 23590107873, "usage": 5, "io": { "inputs": [ { "nodeId": 57, "nodeType": "LoadImage", "file": "sd3.5_large_blur_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_stability_ai_sd3.5_t2i", "title": "Stability AI：SD3.5文生图", "description": "生成具有出色提示遵循性的高质量图像。适用于1百万像素分辨率的专业用例。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "图像", "文生图" ], "models": [ "Stability" ], "date": "2025-03-01", "openSource": false, "size": 0, "vram": 0, "usage": 1, "io": { "inputs": [ { "nodeId": 4, "nodeType": "LoadImage", "file": "example.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 2, "nodeType": "SaveImage", "file": "api_stability_ai_sd3.5_t2i.png", "mediaType": "image" } ] }, "username": "ComfyUI", "logos": [ { "provider": "Stability" } ], "searchRank": 0, "thumbnail": [ "output/api_stability_ai_sd3.5_t2i.png" ] }, { "name": "api_ideogram_v3_t2i", "title": "Ideogram V3：文生图", "description": "使用Ideogram V3生成具有出色提示对齐、照片级真实感和文本渲染的专业品质图像。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "图像", "文生图" ], "models": [ "Ideogram" ], "date": "2025-03-01", "openSource": false, "size": 0, "vram": 0, "usage": 7, "io": { "outputs": [ { "nodeId": 59, "nodeType": "SaveImage", "file": "api_ideogram_v3_t2i.png", "mediaType": "image" } ] }, "username": "ComfyUI", "logos": [ { "provider": "Ideogram" } ], "searchRank": 0, "thumbnail": [ "output/api_ideogram_v3_t2i.png" ] }, { "name": "api_luma_photon_style_ref", "title": "Luma Photon：风格参考", "description": "使用Luma Photon通过混合风格参考并进行精确控制来生成图像。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "图像", "文生图" ], "models": [ "Luma" ], "date": "2025-03-01", "openSource": false, "size": 0, "vram": 0, "usage": 4, "io": { "inputs": [ { "nodeId": 26, "nodeType": "LoadImage", "file": "api_luma_photon_i2i_ref_image_1.png", "mediaType": "image" }, { "nodeId": 52, "nodeType": "LoadImage", "file": "api_luma_photon_i2i_ref_image_2.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 32, "nodeType": "SaveImage", "file": "api_luma_photon_style_ref.png", "mediaType": "image" } ] }, "username": "ComfyUI", "searchRank": 0, "logos": [ { "provider": "Luma" } ], "thumbnail": [ "input/api_luma_photon_i2i_ref_image_1.png", "output/api_luma_photon_style_ref.png" ] }, { "name": "api_krea2_style_reference", "title": "Krea 2：风格参考", "description": "上传风格参考图像并输入文本提示。生成与提示匹配的图像，同时应用参考图像的美学风格。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "风格参考" ], "models": [ "Krea 2" ], "logos": [ { "provider": "Krea" } ], "date": "2026-05-27", "openSource": false, "size": 0, "vram": 0, "usage": 220, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 11, "nodeType": "LoadImage", "file": "krea_style_reference_image1.png", "mediaType": "image" }, { "nodeId": 12, "nodeType": "LoadImage", "file": "krea_style_reference_image2.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 2, "nodeType": "SaveImage", "file": "api_krea2_style_reference.png", "mediaType": "image" } ] } } ] }, { "moduleName": "default", "type": "video", "category": "生成类型", "icon": "icon-[lucide--film]", "title": "视频", "templates": [ { "name": "video_ltx2_3_i2v", "title": "LTX-2.3：图生视频", "description": "上传图像以生成视频，提升运动一致性和精细细节，包括增强的竖屏格式支持和音频质量。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生视频", "视频" ], "models": [ "LTX-2.3", "Lightricks" ], "date": "2026-03-05", "openSource": true, "size": 47244640256, "vram": 47244640256, "usage": 4012, "searchRank": 0, "io": { "inputs": [ { "nodeId": 269, "nodeType": "LoadImage", "file": "egyptian_queen.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_grok_video", "title": "Grok：视频生成", "description": "输入文本提示或初始图像帧，使用 Grok 模型生成最长 15 秒的带同步音频的视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "视频", "图生视频", "文生视频", "API" ], "models": [ "Grok" ], "date": "2026-01-28", "openSource": false, "usage": 2219, "io": { "inputs": [ { "nodeId": 3, "nodeType": "LoadImage", "file": "sci-fi_mech.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 2, "nodeType": "SaveVideo", "file": "api_grok_video.mp4", "mediaType": "video" } ] }, "username": "ComfyUI", "thumbnail": [ "output/api_grok_video.mp4" ] }, { "name": "video_wan2_2_14B_i2v", "title": "Wan 2.2 14B图生视频", "description": "使用Wan 2.2将静态图像转换为具有精确运动控制和风格保持的动态视频。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "hoverDissolve", "tutorialUrl": "https://docs.comfy.org/tutorials/video/wan/wan2_2", "tags": [ "图生视频", "视频" ], "models": [ "Wan", "Wan2.2" ], "date": "2025-07-29", "size": 38010460570, "vram": 38010460570, "usage": 2095, "io": { "inputs": [ { "nodeId": 97, "nodeType": "LoadImage", "file": "video_wan2_2_14B_i2v_input_image.jpg", "mediaType": "image" } ], "outputs": [ { "nodeId": 108, "nodeType": "SaveVideo", "file": "video_wan2_2_14B_i2v.mp4", "mediaType": "video" } ] }, "openSource": true, "searchRank": 0, "username": "ComfyUI", "thumbnail": [ "output/video_wan2_2_14B_i2v.mp4" ] }, { "name": "video_wan2_2_14B_t2v", "title": "Wan 2.2 14B文生视频", "description": "使用Wan 2.2通过电影美学控制和动态运动生成从文本提示生成高质量视频。", "mediaType": "image", "mediaSubtype": "webp", "tutorialUrl": "https://docs.comfy.org/tutorials/video/wan/wan2_2", "tags": [ "文生视频", "视频" ], "models": [ "Wan2.2", "Wan" ], "date": "2025-07-29", "size": 38031935406, "vram": 38031935406, "usage": 265, "username": "ComfyUI" }, { "name": "api_grok_imagine_video_1_5", "title": "Grok Imagine Video 1.5：图像转视频", "description": "上传图片和可选的文字提示，生成带有原生音频、逼真动作和高达720p分辨率的高质量视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生视频" ], "models": [ "Grok" ], "date": "2026-06-01", "openSource": false, "size": 0, "vram": 0, "usage": 0, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 6, "nodeType": "LoadImage", "file": "robot_nun_pray.png", "mediaType": "image" } ] } }, { "name": "video_ltx2_3_ia2v", "title": "LTX-2.3：图像音频转视频", "description": "上传图像和音频文件，使用LTX-2.3生成具有同步唇部动作的高质量视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生视频", "口型同步", "视频" ], "models": [ "LTX-2.3" ], "date": "2026-03-09", "openSource": true, "size": 42949672960, "vram": 42949672960, "usage": 719, "searchRank": 0, "io": { "inputs": [ { "nodeId": 269, "nodeType": "LoadImage", "file": "cactus_man.png", "mediaType": "image" }, { "nodeId": 276, "nodeType": "LoadAudio", "file": "ltx_23_audio.mp3", "mediaType": "audio" } ], "outputs": [ { "nodeId": 341, "nodeType": "SaveVideo", "file": "video_ltx2_3_ia2v.mp4", "mediaType": "video" } ] }, "username": "ComfyUI", "thumbnail": [ "output/video_ltx2_3_ia2v.mp4" ] }, { "name": "api_grok_video_edit", "title": "Grok：视频编辑", "description": "上传视频进行编辑。输入您想要的修改内容。使用Grok API生成编辑后的视频。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "视频编辑" ], "models": [ "Grok" ], "date": "2026-01-28", "openSource": false, "usage": 134, "size": 0, "vram": 0, "searchRank": 0, "io": { "inputs": [ { "nodeId": 8, "nodeType": "LoadVideo", "file": "futuristic_motorcycle.mp4", "mediaType": "video" } ], "outputs": [ { "nodeId": 2, "nodeType": "SaveVideo", "file": "api_grok_video_edit.mp4", "mediaType": "video" } ] }, "username": "ComfyUI", "thumbnail": [ "input/futuristic_motorcycle.mp4", "output/api_grok_video_edit.mp4" ] }, { "name": "video_ltx2_3_ic_lora", "title": "LTX 2.3 IC-LoRA Union Control", "description": "使用对齐的控制输入（如深度、姿态或边缘），通过 IC-LoRA 生成 LTX 2.3 视频。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "ControlNet", "视频" ], "models": [ "LTX-2.3" ], "date": "2026-04-25", "openSource": true, "size": 40909563494, "vram": 40909563494, "usage": 213, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 199, "nodeType": "LoadVideo", "file": "stone_ruins.mp4", "mediaType": "video" }, { "nodeId": 200, "nodeType": "LoadImage", "file": "the_forgotten_gate.png", "mediaType": "image" } ] } }, { "name": "api_grok_reference_to_video", "title": "Grok：视频参考", "description": "上传参考图像并使用Grok API生成视频，最多支持6张参考图像以增强一致性。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "角色参考", "视频" ], "models": [ "Grok" ], "logos": [ { "provider": "Grok" } ], "date": "2026-03-24", "openSource": false, "size": 0, "vram": 0, "usage": 165, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 5, "nodeType": "LoadImage", "file": "quiet_girl.png", "mediaType": "image" }, { "nodeId": 6, "nodeType": "LoadImage", "file": "phone.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 4, "nodeType": "SaveVideo", "file": "api_grok_reference_to_video.mp4", "mediaType": "video" } ] }, "thumbnail": [ "output/api_grok_reference_to_video.mp4" ] }, { "name": "api_grok_video_extend", "title": "Grok：视频扩展", "description": "上传一个短视频和描述下一场景的文本提示。使用 Grok extend 生成无缝的视频扩展。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "视频", "视频扩展" ], "models": [ "Grok" ], "logos": [ { "provider": "Grok" } ], "date": "2026-03-24", "openSource": false, "size": 0, "vram": 0, "usage": 105, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 1, "nodeType": "LoadVideo", "file": "deep_space_nexus.mp4", "mediaType": "video" } ], "outputs": [ { "nodeId": 3, "nodeType": "SaveVideo", "file": "api_grok_video_extend.mp4", "mediaType": "video" } ] }, "thumbnail": [ "output/api_grok_video_extend.mp4" ] }, { "name": "video_ltx2_3_flf2v", "title": "LTX-2.3: FLF2V", "description": "上传起始图像以生成视频序列。该工作流程会创建一个风格一致、带有动感的短视频，从您提供的帧开始。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "首尾帧视频", "视频" ], "models": [ "LTX-2.3" ], "date": "2026-03-16", "openSource": true, "size": 38976828211, "vram": 38976828211, "usage": 688, "searchRank": 0, "io": { "inputs": [ { "nodeId": 31, "nodeType": "LoadImage", "file": "high_view_classic_car.png", "mediaType": "image" }, { "nodeId": 39, "nodeType": "LoadImage", "file": "low_view_classic_car.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 68, "nodeType": "SaveVideo", "file": "video_ltx2_3_flf2v.mp4", "mediaType": "video" } ] }, "thumbnail": [ "output/video_ltx2_3_flf2v.mp4" ], "username": "ComfyUI" }, { "name": "video_ltx2_3_t2v", "title": "LTX-2.3：文生视频", "description": "根据文本提示生成视频，可选择使用参考图像。获得高质量视频，具有改进的运动、音频和细节，优化为纵向或横向格式。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生视频", "视频" ], "models": [ "LTX-2.3", "Lightricks" ], "date": "2026-03-05", "openSource": true, "size": 47244640256, "vram": 47244640256, "usage": 588, "searchRank": 0, "io": { "outputs": [ { "nodeId": 75, "nodeType": "SaveVideo", "file": "video_ltx2_3_t2v.mp4", "mediaType": "video" } ] }, "username": "ComfyUI", "thumbnail": [ "output/video_ltx2_3_t2v.mp4" ] }, { "name": "video_ltx2_3_id_lora", "title": "LTX-2.3: ID LoRA", "description": "通过文本提示、参考图像和简短音频片段生成带同步音频的个性化视频。使用ID-LoRA在单一生成模型中适配人物的外观和声音。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "口型同步", "视频", "语音克隆" ], "models": [ "LTX-2.3" ], "date": "2026-03-27", "openSource": true, "size": 43486543872, "vram": 43486543872, "usage": 209, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 269, "nodeType": "LoadImage", "file": "vintage_thinker.png", "mediaType": "image" }, { "nodeId": 276, "nodeType": "LoadAudio", "file": "ltx23_reference_audio.mp3", "mediaType": "audio" } ], "outputs": [ { "nodeId": 341, "nodeType": "SaveVideo", "file": "video_ltx2_3_id_lora.mp4", "mediaType": "video" } ] }, "thumbnail": [ "output/video_ltx2_3_id_lora.mp4" ] }, { "name": "video_ltx2_i2v", "title": "LTX-2：图生视频", "description": "使用LTX-2将静态图像变为动态视频，音视频同步，表情自然，动作流畅。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生视频", "视频" ], "models": [ "LTX-2", "Lightricks" ], "date": "2026-01-06", "size": 38010460570, "vram": 38010460570, "usage": 246, "searchRank": 0, "openSource": true, "io": { "inputs": [ { "nodeId": 98, "nodeType": "LoadImage", "file": "freckled_curly_girl.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 75, "nodeType": "SaveVideo", "file": "video_ltx2_i2v.mp4", "mediaType": "video" } ] }, "username": "ComfyUI", "thumbnail": [ "output/video_ltx2_i2v.mp4" ] }, { "name": "video_ltx2_t2v", "title": "LTX-2 文生视频", "description": "使用LTX-2将文本提示生成高质量视频，音视频同步，支持表情口型和动态动作。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生视频", "视频" ], "models": [ "LTX-2", "Lightricks" ], "date": "2026-01-06", "size": 45526653338, "vram": 45526653338, "usage": 53, "searchRank": 0, "openSource": true, "io": { "outputs": [ { "nodeId": 75, "nodeType": "SaveVideo", "file": "video_ltx2_t2v.mp4", "mediaType": "video" } ] }, "username": "ComfyUI", "thumbnail": [ "output/video_ltx2_t2v.mp4" ] }, { "name": "api_happyhorse1_0_i2v", "title": "HappyHorse：图片转视频", "description": "使用HappyHorse-1.0-I2V从图像生成高质量、流畅的视频，具备逼真的动态渲染和准确的文本与图像语义理解。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生视频" ], "models": [ "HappyHorse 1.0" ], "logos": [ { "provider": "HappyHorse" } ], "date": "2026-04-27", "openSource": false, "size": 0, "vram": 0, "usage": 150, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 14, "nodeType": "LoadImage", "file": "pink_city_horse.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 6, "nodeType": "SaveVideo", "file": "api_happyhorse1_0_i2v.mp4", "mediaType": "video" } ] }, "thumbnail": [ "output/api_happyhorse1_0_i2v.mp4" ] }, { "name": "api_happyhorse1_0_t2v", "title": "HappyHorse：文字转视频", "description": "通过文本提示生成高质量、逼真的视频，具有流畅的动作和丰富的细节。输入文本描述，以产生具有准确语义理解的动态视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "文生视频" ], "models": [ "HappyHorse 1.0" ], "logos": [ { "provider": "HappyHorse" } ], "date": "2026-04-27", "openSource": false, "size": 0, "vram": 0, "usage": 26, "searchRank": 0, "username": "ComfyUI", "io": { "outputs": [ { "nodeId": 6, "nodeType": "SaveVideo", "file": "api_happyhorse1_0_t2v.mp4", "mediaType": "video" } ] }, "thumbnail": [ "output/api_happyhorse1_0_t2v.mp4" ] }, { "name": "api_happyhorse1_0_video_edit", "title": "HappyHorse：视频编辑", "description": "使用文本指令和可选的参考图像进行风格迁移或局部替换来编辑视频。输入一个视频和最多5张参考图像，输出编辑后的视频。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "视频编辑", "API" ], "models": [ "HappyHorse 1.0" ], "logos": [ { "provider": "HappyHorse" } ], "date": "2026-04-27", "openSource": false, "size": 0, "vram": 0, "usage": 211, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 14, "nodeType": "LoadImage", "file": "urban_man.png", "mediaType": "image" }, { "nodeId": 16, "nodeType": "LoadVideo", "file": "woman_by_tram.mp4", "mediaType": "video" } ], "outputs": [ { "nodeId": 6, "nodeType": "SaveVideo", "file": "api_happyhorse1_0_video_edit.mp4", "mediaType": "video" } ] }, "thumbnail": [ "input/woman_by_tram.mp4", "output/api_happyhorse1_0_video_edit.mp4" ] }, { "name": "api_happyhorse1_0_r2v", "title": "HappyHorse：视频参考", "description": "使用文本提示，从多个参考图像中生成保留主体角色的视频。输入最多9张参考图像和场景描述，生成3-15秒的视频输出。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "视频参考", "API" ], "models": [ "HappyHorse 1.0" ], "logos": [ { "provider": "HappyHorse" } ], "date": "2026-04-27", "openSource": false, "size": 0, "vram": 0, "usage": 60, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 14, "nodeType": "LoadImage", "file": "car_ref.png", "mediaType": "image" }, { "nodeId": 18, "nodeType": "LoadImage", "file": "man_ref.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 6, "nodeType": "SaveVideo", "file": "api_happyhorse1_0_r2v.mp4", "mediaType": "video" } ] }, "thumbnail": [ "output/api_happyhorse1_0_r2v.mp4" ] }, { "name": "api_wan2_7_i2v", "title": "Wan2.7：图生视频", "description": "上传图像和音频文件以生成带同步声音的视频，使用首尾帧进行控制。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生视频", "视频" ], "models": [ "Wan2.7" ], "logos": [ { "provider": "WAN" } ], "date": "2026-04-02", "openSource": false, "size": 0, "vram": 0, "usage": 405, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 24, "nodeType": "LoadImage", "file": "tv_head.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 32, "nodeType": "SaveVideo", "file": "api_wan2_7_i2v.mp4", "mediaType": "video" } ] } }, { "name": "api_kling_motion_control3", "title": "Kling 3.0: Motion Control", "description": "上传角色图像和动作参考视频。生成新视频，让您的角色执行参考视频中的动作，并通过可选元素绑定增强面部一致性。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "运动控制" ], "models": [ "Kling 3.0" ], "logos": [ { "provider": "Kling" } ], "date": "2026-03-06", "openSource": false, "size": 0, "vram": 0, "usage": 288, "searchRank": 0, "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadImage", "file": "future_robot.png", "mediaType": "image" }, { "nodeId": 3, "nodeType": "LoadVideo", "file": "boxing_man.mp4", "mediaType": "video" } ] }, "username": "ComfyUI" }, { "name": "api_kling_v3_video", "title": "Kling 3.0：视频生成", "description": "输入文本或图像提示，生成具有多镜头构图、复杂摄像机运动和一致主体的15秒视频序列。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "图生视频", "视频" ], "models": [ "Kling 3.0" ], "logos": [ { "provider": "Kling" } ], "date": "2026-02-10", "openSource": false, "size": 0, "vram": 0, "usage": 827, "searchRank": 0, "io": { "inputs": [ { "nodeId": 1, "nodeType": "LoadImage", "file": "boxing.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 2, "nodeType": "SaveVideo", "file": "api_kling_v3_video.mp4", "mediaType": "video" } ] }, "username": "ComfyUI", "thumbnail": [ "output/api_kling_v3_video.mp4" ] }, { "name": "api_bytedance_image_to_video", "title": "ByteDance: 图生视频", "description": "使用 ByteDance 的 Seedance 模型将静态图像转换为动态视频。分析图像结构并生成自然运动，具有一致的视觉风格和连贯的视频序列。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "图生视频", "视频" ], "models": [ "Seedance1.0 Pro", "ByteDance" ], "date": "2025-10-06", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 937, "io": { "inputs": [ { "nodeId": 12, "nodeType": "LoadImage", "file": "api_bytedance_image_to_video_input_image.png", "mediaType": "image" } ] }, "logos": [ { "provider": "ByteDance" } ], "searchRank": 0, "username": "ComfyUI" }, { "name": "api_bytedance_seedance1_5_image_to_video", "title": "Seedance 1.5 Pro：图生视频", "description": "使用seedance1.5 Pro将图像转化为具有语音、动作和同步音效的鲜活角色。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "图生视频" ], "models": [ "Seedance" ], "date": "2026-01-19", "openSource": false, "size": 0, "vram": 0, "usage": 790, "searchRank": 0, "io": { "inputs": [ { "nodeId": 4, "nodeType": "LoadImage", "file": "young_baseball_player.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 9, "nodeType": "SaveVideo", "file": "api_bytedance_seedance1_5_image_to_video.mp4", "mediaType": "video" } ] }, "username": "ComfyUI", "logos": [ { "provider": "ByteDance" } ], "thumbnail": [ "output/api_bytedance_seedance1_5_image_to_video.mp4" ] }, { "name": "api_wan2_7_video_edit", "title": "Wan2.7：视频编辑", "description": "上传视频及对应图像，使用Wan2.7模型编辑并替换角色或场景。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "视频", "视频编辑" ], "models": [ "Wan2.7" ], "logos": [ { "provider": "WAN" } ], "date": "2026-04-02", "openSource": false, "size": 0, "vram": 0, "usage": 238, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 1, "nodeType": "LoadVideo", "file": "play_chess.mp4", "mediaType": "video" }, { "nodeId": 2, "nodeType": "LoadImage", "file": "old_bear.png", "mediaType": "image" }, { "nodeId": 3, "nodeType": "LoadImage", "file": "funny_dog.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 5, "nodeType": "SaveVideo", "file": "api_wan2_7_video_edit.mp4", "mediaType": "video" } ] } }, { "name": "api_wan2_7_t2v", "title": "Wan2.7：文生视频", "description": "使用Wan 2.7从文本提示生成视频。支持音频参考以创建口型同步视频", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "口型同步", "文生视频", "视频" ], "models": [ "Wan2.7" ], "date": "2026-04-02", "openSource": false, "size": 0, "vram": 0, "usage": 44, "searchRank": 0, "username": "ComfyUI", "logos": [ { "provider": "WAN" } ] }, { "name": "api_wan2_7_r2v", "title": "Wan2.7：视频参考", "description": "上传参考视频和角色图像，生成具有一致角色特征的视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "视频" ], "models": [ "Wan2.7" ], "logos": [ { "provider": "WAN" } ], "date": "2026-04-02", "openSource": false, "size": 0, "vram": 0, "usage": 51, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 4, "nodeType": "LoadImage", "file": "steampunk_woman_underwater.png", "mediaType": "image" }, { "nodeId": 5, "nodeType": "LoadImage", "file": "x_robot.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 7, "nodeType": "SaveVideo", "file": "api_wan2_7_r2v.mp4", "mediaType": "video" } ] } }, { "name": "api_kling_avatar2", "title": "Kling：Avatar 2.0", "description": "上传一张肖像和一个音频文件，生成具有同步唇形和自然面部表情的说话头像视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "口型同步", "API" ], "models": [ "Kling" ], "logos": [ { "provider": "Kling" } ], "date": "2026-02-26", "openSource": false, "size": 0, "vram": 0, "usage": 48, "searchRank": 0, "io": { "inputs": [ { "nodeId": 3, "nodeType": "LoadImage", "file": "woman_in_red.png", "mediaType": "image" }, { "nodeId": 4, "nodeType": "LoadAudio", "file": "kling_avatar_input.mp3", "mediaType": "audio" } ], "outputs": [ { "nodeId": 5, "nodeType": "SaveVideo", "file": "api_kling_avatar2.mp4", "mediaType": "video" } ] }, "username": "ComfyUI", "thumbnail": [ "output/api_kling_avatar2.mp4" ] }, { "name": "video_wan_vace_14B_v2v", "title": "Wan VACE控制视频", "description": "使用Wan VACE通过控制输入视频和参考图像生成视频。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tutorialUrl": "https://docs.comfy.org/tutorials/video/wan/vace", "tags": [ "视频转视频", "视频" ], "models": [ "Wan2.1", "Wan" ], "date": "2025-05-21", "size": 57756572713, "vram": 57756572713, "usage": 106, "io": { "inputs": [ { "nodeId": 134, "nodeType": "LoadImage", "file": "video_wan_vace_14B_v2v_reference_image.jpg", "mediaType": "image" }, { "nodeId": 145, "nodeType": "LoadVideo", "file": "video_wan_vace_14B_v2v_reference_image_control_video.mp4", "mediaType": "video" } ], "outputs": [ { "nodeId": 114, "nodeType": "SaveVideo", "file": "video_wan_vace_14B_v2v.mp4", "mediaType": "video" } ] }, "username": "ComfyUI" }, { "name": "api_veo3", "title": "Veo3：图生视频", "description": "使用Google的高级Veo 3 API从文本提示或图像生成高质量8秒视频。具有音频生成、提示增强和速度或质量的双重模型选项。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生视频", "文生视频", "API" ], "models": [ "Veo", "Google" ], "date": "2025-03-01", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 338, "io": { "inputs": [ { "nodeId": 11, "nodeType": "LoadImage", "file": "api_veo3_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_kling_v3_flf2v", "title": "Kling 3.0：首尾帧视频", "description": "输入首尾帧图像，生成连续视频序列，支持多镜头生成、精确元素控制及多语言提示。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "视频", "首尾帧视频" ], "models": [ "Kling 3.0" ], "logos": [ { "provider": "Kling" } ], "date": "2026-02-10", "openSource": false, "size": 0, "vram": 0, "usage": 447, "searchRank": 0, "io": { "inputs": [ { "nodeId": 1, "nodeType": "LoadImage", "file": "goldenfish_in_bag.png", "mediaType": "image" }, { "nodeId": 13, "nodeType": "LoadImage", "file": "man_in_daydream.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 2, "nodeType": "SaveVideo", "file": "Kling_v3_flf2v.mp4", "mediaType": "video" } ] }, "username": "ComfyUI", "thumbnail": [ "output/Kling_v3_flf2v.mp4" ] }, { "name": "video_causal_forcing_i2v", "title": "因果驱动：图像到视频", "description": "输入文本提示并选择模型检查点。使用因果强制或因果强制++生成高质量视频，推理步骤为1到4步。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生视频" ], "models": [ "Wan2.1" ], "date": "2026-03-31", "openSource": true, "size": 12670153523, "vram": 12670153523, "usage": 0, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 18, "nodeType": "LoadImage", "file": "girl_riding_a_wolf.png", "mediaType": "image" } ] } }, { "name": "api_kling_o3_video_edit", "title": "Kling O3：视频编辑", "description": "上传视频和参考图片，使用Kling O3进行编辑。生成专业编辑的视频，具备精确的主体编辑、场景构图和原生音视频同步。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "视频编辑", "API", "视频" ], "models": [ "Kling O3" ], "logos": [ { "provider": "Kling" } ], "date": "2026-02-10", "openSource": false, "size": 0, "vram": 0, "usage": 152, "searchRank": 0, "io": { "inputs": [ { "nodeId": 17, "nodeType": "LoadImage", "file": "giant_trees_sky_pyramid.png", "mediaType": "image" }, { "nodeId": 25, "nodeType": "LoadVideo", "file": "kling_o3_video_input.mp4", "mediaType": "video" } ] }, "username": "ComfyUI" }, { "name": "api_kling_o3_flf2v", "title": "Kling O3：首尾帧视频", "description": "上传参考图像或指定起始帧和结束帧。生成具有精确语义控制、更长时长和更强叙事连贯性的视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "首尾帧视频" ], "models": [ "Kling O3" ], "logos": [ { "provider": "Kling" } ], "date": "2026-02-10", "openSource": false, "size": 0, "vram": 0, "usage": 68, "searchRank": 0, "io": { "inputs": [ { "nodeId": 17, "nodeType": "LoadImage", "file": "ivory_rider_right_front_view.png", "mediaType": "image" }, { "nodeId": 34, "nodeType": "LoadImage", "file": "ivory_rider.png", "mediaType": "image" }, { "nodeId": 35, "nodeType": "LoadImage", "file": "ivory_rider_left_side_view.png", "mediaType": "image" }, { "nodeId": 36, "nodeType": "LoadImage", "file": "ivory_rider_right_side_view.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_king_o3_t2v", "title": "Kling O3：文生视频", "description": "上传文本、音频或图像，生成具有同步音频、角色一致性和故事板控制的15秒视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "文生视频" ], "models": [ "Kling O3" ], "logos": [ { "provider": "Kling" } ], "date": "2026-02-10", "openSource": false, "size": 0, "vram": 0, "usage": 41, "searchRank": 0, "io": { "outputs": [ { "nodeId": 14, "nodeType": "SaveVideo", "file": "api_king_o3_t2v.mp4", "mediaType": "video" } ] }, "username": "ComfyUI", "thumbnail": [ "output/api_king_o3_t2v.mp4" ] }, { "name": "api_kling_omni_v2v", "title": "Kling O1: 视频转视频", "description": "使用 Kling O1 转换视频。通过文本提示和参考图像编辑视频内容、更改风格或替换角色。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "hoverDissolve", "tags": [ "API", "角色参考", "视频" ], "models": [ "Kling O1" ], "date": "2026-01-10", "openSource": false, "usage": 47, "size": 0, "vram": 0, "searchRank": 0, "io": { "inputs": [ { "nodeId": 3, "nodeType": "LoadImage", "file": "blue_shirt_rabbit.jpg", "mediaType": "image" }, { "nodeId": 7, "nodeType": "LoadVideo", "file": "dancing_women.mp4", "mediaType": "video" } ] }, "username": "ComfyUI" }, { "name": "video_ltx2_i2v_lora", "title": "使用LTX-2 I2V和LoRA压缩一切", "description": "上传任意图像以应用Squish It LoRA效果。使用LTX-2 I2V模型生成挤压动画。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生视频" ], "models": [ "LTX-2" ], "date": "2026-02-09", "openSource": true, "size": 103830834381, "vram": 103830834381, "usage": 25, "searchRank": 0, "io": { "inputs": [ { "nodeId": 98, "nodeType": "LoadImage", "file": "puppy_with_blue_bandana.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 75, "nodeType": "SaveVideo", "file": "video_ltx2_i2v_lora.mp4", "mediaType": "video" } ] }, "username": "ComfyUI", "thumbnail": [ "output/video_ltx2_i2v_lora.mp4" ] }, { "name": "api_kling_o3_i2v", "title": "Kling O3：视频参考", "description": "上传参考图像及可选的音频或文本提示。生成角色一致的视频，具备原生音频输出和精确的故事板控制。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "角色参考", "图生视频" ], "models": [ "Kling O3" ], "logos": [ { "provider": "Kling" } ], "date": "2026-02-10", "openSource": false, "size": 0, "vram": 0, "usage": 154, "searchRank": 0, "io": { "inputs": [ { "nodeId": 17, "nodeType": "LoadImage", "file": "chasing_the_light.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 20, "nodeType": "SaveVideo", "file": "kling_o3_i2v.mp4", "mediaType": "video" } ] }, "username": "ComfyUI", "thumbnail": [ "output/kling_o3_i2v.mp4" ] }, { "name": "api_bytedance_seedance1_5_flf2v", "title": "Seedance 1.5 Pro: FLF2V", "description": "上传起始帧和结束帧，Seedance 1.5 Pro 会自动在它们之间生成带有同步音频的流畅视频过渡。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "首尾帧视频", "图生视频", "API" ], "models": [ "ByteDance", "Seedance" ], "date": "2026-01-19", "openSource": false, "size": 0, "vram": 0, "usage": 183, "searchRank": 0, "io": { "inputs": [ { "nodeId": 4, "nodeType": "LoadImage", "file": "dancing_capybara.png", "mediaType": "image" }, { "nodeId": 11, "nodeType": "LoadImage", "file": "dj_capybara.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "video_wan2_2_14B_flf2v", "title": "Wan 2.2 14B首尾帧到视频", "description": "通过定义开始和结束帧生成平滑的视频过渡。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "hoverDissolve", "tutorialUrl": "https://docs.comfy.org/tutorials/video/wan/wan2_2", "tags": [ "首尾帧视频", "视频" ], "models": [ "Wan2.2", "Wan" ], "date": "2025-08-02", "size": 38031935406, "vram": 38031935406, "usage": 149, "io": { "inputs": [ { "nodeId": 62, "nodeType": "LoadImage", "file": "video_wan2_2_14B_flf2v_start_image.png", "mediaType": "image" }, { "nodeId": 68, "nodeType": "LoadImage", "file": "video_wan2_2_14B_flf2v_end_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "video_wan2_2_14B_fun_control", "title": "Wan 2.2 14B 趣味控制", "description": "使用Wan 2.2 Fun Control 通过姿态、深度和边缘控制引导生成视频。", "mediaType": "image", "mediaSubtype": "webp", "tutorialUrl": "https://docs.comfy.org/tutorials/video/wan/wan2-2-fun-control", "tags": [ "视频转视频", "视频", "ControlNet" ], "models": [ "Wan2.2", "Wan" ], "date": "2025-08-12", "size": 38031935406, "vram": 38031935406, "usage": 69, "io": { "inputs": [ { "nodeId": 145, "nodeType": "LoadImage", "file": "video_wan2_2_14B_fun_control_start_image.jpg", "mediaType": "image" }, { "nodeId": 158, "nodeType": "LoadVideo", "file": "video_wan2_2_14B_fun_control_start_image_control_video.mp4", "mediaType": "video" } ] }, "username": "ComfyUI" }, { "name": "api_vidu_q3_text_to_video", "title": "Vidu Q3 视听视频生成", "description": "输入文本提示或上传图像，生成1-16秒带同步音频和自动场景转换的视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "文生视频", "视频" ], "models": [ "Vidu", "Vidu Q3" ], "date": "2026-02-02", "openSource": false, "size": 0, "vram": 0, "usage": 3, "searchRank": 0, "username": "ComfyUI" }, { "name": "api_vidu_q3_image_to_video", "title": "Vidu Q3 图生视频带音频", "description": "上传一张图片并提供文本提示，即可生成一段1-16秒的视频，包含同步音频和智能场景转换。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "图生视频", "视频" ], "models": [ "Vidu", "Vidu Q3" ], "date": "2026-02-02", "openSource": false, "usage": 69, "size": 0, "vram": 0, "logos": [ { "provider": "Vidu" } ], "searchRank": 0, "io": { "inputs": [ { "nodeId": 16, "nodeType": "LoadImage", "file": "two_characters.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_wan2_6_i2v", "title": "Wan2.6: 图生视频", "description": "将图像转换为高质量视频，具备增强的图像质量、更流畅的运动、1080P分辨率支持和自然的运动生成，实现专业效果。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生视频", "视频", "API" ], "models": [ "Wan2.6", "Wan" ], "date": "2025-12-20", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 169, "io": { "inputs": [ { "nodeId": 42, "nodeType": "LoadImage", "file": "red_car.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_kling_motion_control", "title": "Kling2.6：运动控制", "description": "通过同步运动控制，将参考视频中的精确角色动作和表情应用到您的角色图像上。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "视频", "运动控制", "API" ], "models": [ "Kling 2.6", "Kling" ], "date": "2026-01-12", "openSource": false, "usage": 249, "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadVideo", "file": "street_dancer.mp4", "mediaType": "video" }, { "nodeId": 11, "nodeType": "LoadImage", "file": "streetwear_fox.jpg", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_kling2_6_i2v", "title": "Kling2.6：图生视频（音画同步）", "description": "将静态图片转为带对话、演唱、音效和环境音的动态视频", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生视频", "视频", "API", "音频" ], "models": [ "Kling 2.6", "Kling" ], "date": "2025-12-22", "openSource": false, "size": 0, "vram": 0, "usage": 40, "io": { "inputs": [ { "nodeId": 14, "nodeType": "LoadImage", "file": "lizard_rapper_in_bar.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_vidu_q2_i2v", "title": "Vidu Q2: 图生视频", "description": "将静态图像转换为动态1080p视频，具有电影级相机控制和多主体一致性。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生视频", "API" ], "models": [ "Vidu Q2", "Vidu" ], "date": "2026-01-12", "openSource": false, "usage": 120, "io": { "inputs": [ { "nodeId": 40, "nodeType": "LoadImage", "file": "cozy_bear.jpg", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "video_ltx2_i2v_distilled", "title": "LTX-2 图生视频（蒸馏版）", "description": "使用LTX-2蒸馏模型将静态图像变为动态视频，音视频同步。专为快速生成优化，同时保持质量。支持表情口型、自然动作，生成速度更快。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生视频", "视频" ], "models": [ "LTX-2", "Lightricks" ], "date": "2026-01-06", "size": 38010460570, "vram": 38010460570, "usage": 99, "searchRank": 0, "openSource": true, "io": { "inputs": [ { "nodeId": 98, "nodeType": "LoadImage", "file": "snow_bike_jump_crew.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "video_ltx2_t2v_distilled", "title": "LTX-2 文生视频（蒸馏版）", "description": "使用LTX-2蒸馏模型将文本提示生成高质量视频，音视频同步。专为快速生成优化，同时保持质量。支持表情口型、动态动作，生成速度更快。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生视频", "视频" ], "models": [ "LTX-2", "Lightricks" ], "date": "2026-01-06", "size": 38010460570, "vram": 38010460570, "usage": 28, "searchRank": 0, "openSource": true, "username": "ComfyUI" }, { "name": "video_ltx2_canny_to_video", "title": "LTX-2 边缘检测图生视频", "description": "使用LTX-2通过边缘检测（Canny）引导生成高质量视频，音视频同步，边缘控制精准，结构一致性好。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "hoverDissolve", "tags": [ "ControlNet", "视频" ], "models": [ "LTX-2", "Lightricks" ], "date": "2026-01-06", "size": 38010460570, "vram": 38010460570, "usage": 13, "searchRank": 0, "openSource": true, "io": { "inputs": [ { "nodeId": 127, "nodeType": "LoadVideo", "file": "squirrel_in_flower_garden.mp4", "mediaType": "video" }, { "nodeId": 160, "nodeType": "LoadImage", "file": "ltx_2_canny_1st_frame.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "video_ltx2_depth_to_video", "title": "LTX-2 深度转视频", "description": "使用LTX-2通过深度图生成高质量视频，音视频同步、精准深度控制、空间感知。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "ControlNet", "视频" ], "thumbnailVariant": "hoverDissolve", "models": [ "LTX-2", "Lightricks" ], "date": "2026-01-06", "size": 38010460570, "vram": 38010460570, "usage": 26, "searchRank": 0, "openSource": true, "io": { "inputs": [ { "nodeId": 160, "nodeType": "LoadImage", "file": "minimalist_sky_view_lounge.png", "mediaType": "image" }, { "nodeId": 183, "nodeType": "LoadVideo", "file": "minimalist_sky_view_lounge.mp4", "mediaType": "video" } ] }, "username": "ComfyUI" }, { "name": "video_hunyuan_video_1.5_720p_i2v", "title": "Hunyuan Video 1.5 图生视频", "description": "将静态图像动画化为动态视频，具备精确的运动与镜头控制。在带来流畅自然动作的同时，还能保持照片和插画的视觉一致性并赋予其生命力。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生视频", "视频" ], "models": [ "Hunyuan Video" ], "date": "2025-11-21", "size": 45384919416, "vram": 45384919416, "usage": 96, "io": { "inputs": [ { "nodeId": 80, "nodeType": "LoadImage", "file": "video_hunyuan_video_1.5_720p_i2v_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_kling_omni_edit_video", "title": "Kling O1：视频编辑", "description": "支持自然语言指令编辑视频，新增视频参考模式，可快速完成风格转换、元素添加、背景修改等多种高质量视频创作任务。", "mediaType": "image", "thumbnailVariant": "compareSlider", "mediaSubtype": "webp", "tags": [ "视频", "API", "视频编辑", "文生视频", "图生视频" ], "models": [ "Kling" ], "date": "2025-12-02", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 102, "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadImage", "file": "api_flux2_input_image_4.png", "mediaType": "image" }, { "nodeId": 11, "nodeType": "LoadImage", "file": "api_flux2_input_image_1.png", "mediaType": "image" }, { "nodeId": 13, "nodeType": "LoadImage", "file": "api_flux2_input_image_2.png", "mediaType": "image" }, { "nodeId": 14, "nodeType": "LoadImage", "file": "api_flux2_input_image_3.png", "mediaType": "image" }, { "nodeId": 23, "nodeType": "LoadVideo", "file": "api_kling_omni_edit_video_input_video.mp4", "mediaType": "video" }, { "nodeId": 24, "nodeType": "LoadImage", "file": "api_kling_omni_edit_video_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_kling_omni_i2v", "title": "Kling O1: 图生视频", "description": "使用 Kling O1 将静态图像转换为动态视频。通过文本提示为图像添加运动、相机移动和生命力。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "视频", "角色参考" ], "models": [ "Kling O1" ], "date": "2026-01-12", "openSource": false, "usage": 37, "io": { "inputs": [ { "nodeId": 3, "nodeType": "LoadImage", "file": "sheep_knit.jpg", "mediaType": "image" }, { "nodeId": 5, "nodeType": "LoadImage", "file": "quokka_bath.jpg", "mediaType": "image" }, { "nodeId": 6, "nodeType": "LoadImage", "file": "quokka_relax.jpg", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_wan_image_to_video", "title": "Wan2.5: 图生视频", "description": "将图像转换为具有同步音频、增强运动和卓越质量的视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生视频", "视频", "API" ], "models": [ "Wan2.5", "Wan" ], "date": "2025-09-27", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 123, "io": { "inputs": [ { "nodeId": 3, "nodeType": "LoadAudio", "file": "audio_ace_step_1_m2m_editing_input_audio.mp3", "mediaType": "audio" }, { "nodeId": 11, "nodeType": "LoadImage", "file": "api_wan_image_to_video_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_kling_flf", "title": "Kling: 首尾帧", "description": "通过控制首尾帧生成视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "视频", "API", "首尾帧视频" ], "models": [ "Kling", "Kling 1.6" ], "date": "2025-03-01", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 54, "io": { "inputs": [ { "nodeId": 40, "nodeType": "LoadImage", "file": "api_kling_flf_first_frame.png", "mediaType": "image" }, { "nodeId": 42, "nodeType": "LoadImage", "file": "api_kling_flf_last_frame.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_kling_i2v", "title": "Kling：图生视频", "description": "使用Kling生成具有良好提示遵循性的视频，适用于动作、表情和摄像机移动。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生视频", "视频", "API" ], "models": [ "Kling", "Kling 2.0" ], "date": "2025-03-01", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 112, "io": { "inputs": [ { "nodeId": 40, "nodeType": "LoadImage", "file": "api_kling_i2v_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_openai_sora_video", "title": "Sora 2: 视频生成", "description": "OpenAI 的 Sora-2 和 Sora-2 Pro 视频生成，具有同步音频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生视频", "文生视频", "API" ], "models": [ "OpenAI" ], "date": "2025-10-08", "openSource": false, "size": 0, "vram": 0, "usage": 68, "io": { "inputs": [ { "nodeId": 12, "nodeType": "LoadImage", "file": "api_openai_sora_video_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_vidu_q2_flf2v", "title": "Vidu Q2: FLF2V", "description": "在首尾帧间生成流畅视频过渡，支持电影感镜头和多主体一致。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "首尾帧视频", "视频", "API" ], "models": [ "Vidu Q2" ], "date": "2026-01-12", "openSource": false, "usage": 29, "io": { "inputs": [ { "nodeId": 40, "nodeType": "LoadImage", "file": "nun_in_church-1.png", "mediaType": "image" }, { "nodeId": 42, "nodeType": "LoadImage", "file": "nun_in_church-2.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "video_wan2_2_5B_ti2v", "title": "Wan 2.2 5B视频生成", "description": "使用Wan 2.2 5B混合模型从文本或图像生成视频", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生视频", "视频" ], "models": [ "Wan2.2", "Wan" ], "date": "2025-07-29", "size": 18146236826, "vram": 18146236826, "usage": 73, "io": { "inputs": [ { "nodeId": 56, "nodeType": "LoadImage", "file": "example.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "video_wan2_2_14B_s2v", "title": "Wan2.2-S2V 音频驱动视频生成", "description": "将静态图像和音频转换为动态视频，具有完美的同步和分钟级生成。", "mediaType": "image", "mediaSubtype": "webp", "tutorialUrl": "https://docs.comfy.org/tutorials/video/wan/wan2-2-s2v", "tags": [ "视频" ], "models": [ "Wan2.2", "Wan" ], "date": "2025-08-02", "size": 25254407700, "vram": 25254407700, "usage": 44, "io": { "inputs": [ { "nodeId": 52, "nodeType": "LoadImage", "file": "video_wan2_2_14B_s2v_reference_image.jpg", "mediaType": "image" }, { "nodeId": 58, "nodeType": "LoadAudio", "file": "video_wan2_2_14B_s2v_input_audio.MP3", "mediaType": "audio" } ] }, "username": "ComfyUI" }, { "name": "video_wan_vace_inpainting", "title": "Wan VACE修复", "description": "在保留周围内容的同时编辑视频中的特定区域。非常适合对象移除或替换。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tutorialUrl": "https://docs.comfy.org/tutorials/video/wan/vace", "tags": [ "图像修复", "视频" ], "models": [ "Wan2.1", "Wan" ], "date": "2025-05-21", "size": 57767310131, "vram": 57767310131, "usage": 65, "io": { "inputs": [ { "nodeId": 162, "nodeType": "LoadImage", "file": "api_bfl_flux_1_kontext_multiple_images_input_dog.jpg", "mediaType": "image" }, { "nodeId": 209, "nodeType": "LoadVideo", "file": "video_wan_vace_inpainting_input_video.mp4", "mediaType": "video" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0 }, { "name": "api_bytedance_flf2v", "title": "ByteDance: 起始结束到视频", "description": "使用 ByteDance 的 Seedance 模型生成电影级视频过渡，具有流畅的运动、场景一致性和专业打磨。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "首尾帧视频", "视频" ], "models": [ "Seedance1.0 Pro", "ByteDance" ], "date": "2025-10-06", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 44, "io": { "inputs": [ { "nodeId": 35, "nodeType": "LoadImage", "file": "api_bytedance_flf2v_first_frame.png", "mediaType": "image" }, { "nodeId": 36, "nodeType": "LoadImage", "file": "api_bytedance_flf2v_last_frame.png", "mediaType": "image" } ] }, "searchRank": 0, "username": "ComfyUI" }, { "name": "api_vidu_q2_r2v", "title": "Vidu Q2：参考生视频", "description": "生成具有参考一致性的视频，最多支持3个主体，在整个视频序列中保持角色和风格的连续性，并具有电影级相机运动。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "角色参考", "图生视频", "API" ], "models": [ "Vidu Q2", "Vidu" ], "date": "2026-01-12", "openSource": false, "usage": 24, "io": { "inputs": [ { "nodeId": 40, "nodeType": "LoadImage", "file": "api_vidu_q2_r2v_red_haired_girl.jpg", "mediaType": "image" }, { "nodeId": 42, "nodeType": "LoadImage", "file": "vintage_robot.jpg", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_kling_omni_t2v", "title": "Kling O1：文生视频", "description": "用 Kling O1 文本生成视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "文生视频", "视频" ], "models": [ "Kling O1" ], "date": "2026-01-12", "openSource": false, "usage": 13, "username": "ComfyUI" }, { "name": "api_ltxv_image_to_video", "title": "LTX-2：图生视频", "description": "使用 LTX-2 Pro 将静态图像转化为动态视频。可生成具有自然运动、同步音频且支持最高 4K@50fps 的电影级视频序列。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生视频", "视频", "API" ], "models": [ "LTX-2", "Lightricks" ], "date": "2025-10-28", "openSource": false, "size": 0, "vram": 0, "usage": 13, "io": { "inputs": [ { "nodeId": 47, "nodeType": "LoadImage", "file": "api_ltxv_image_to_video_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "video_wan2_2_14B_fun_inpaint", "title": "Wan 2.2 14B 趣味修复", "description": "使用 Wan 2.2 Fun Inp 从开始和结束帧生成视频。", "mediaType": "image", "mediaSubtype": "webp", "tutorialUrl": "https://docs.comfy.org/tutorials/video/wan/wan2-2-fun-inp", "tags": [ "首尾帧视频", "视频" ], "models": [ "Wan2.2", "Wan" ], "date": "2025-08-12", "size": 38031935406, "vram": 38031935406, "usage": 46, "io": { "inputs": [ { "nodeId": 110, "nodeType": "LoadImage", "file": "video_wan2_2_14B_fun_inpaint_start_image.png", "mediaType": "image" }, { "nodeId": 112, "nodeType": "LoadImage", "file": "video_wan2_2_14B_fun_inpaint_end_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_runway_gen4_turo_image_to_video", "title": "Runway：Gen4 Turbo图生视频", "description": "使用Runway Gen4 Turbo从图像生成动态视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生视频", "视频", "API" ], "models": [ "Runway" ], "date": "2025-03-01", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 25, "io": { "inputs": [ { "nodeId": 4, "nodeType": "LoadImage", "file": "api_runway_gen4_turo_image_to_video_input_image.jpg", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "ltxv_image_to_video", "title": "LTXV图生视频", "mediaType": "image", "mediaSubtype": "webp", "description": "从静态图像生成视频。", "tutorialUrl": "https://docs.comfy.org/tutorials/video/ltxv", "tags": [ "图生视频", "视频" ], "models": [ "LTX-0.9.5" ], "date": "2025-03-01", "size": 19155554140, "vram": 19155554140, "usage": 69, "io": { "inputs": [ { "nodeId": 78, "nodeType": "LoadImage", "file": "ltxv_image_to_video_input_image.jpg", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_wan2_6_t2v", "title": "Wan2.5: 文生视频", "description": "生成具有同步音频、增强运动和卓越质量的视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生视频", "视频", "API" ], "models": [ "Wan2.6", "Wan" ], "date": "2025-12-20", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 21, "username": "ComfyUI" }, { "name": "video_hunyuan_video_1.5_720p_t2v", "title": "Hunyuan Video 1.5 文生视频", "description": "由文本提示生成高质量 720p 视频，具备电影级镜头控制、情感表达和物理仿真。支持多种风格，包括写实、动漫和 3D 文字渲染。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生视频", "视频" ], "models": [ "Hunyuan Video" ], "date": "2025-11-21", "size": 45419279155, "vram": 45419279155, "usage": 14, "io": { "outputs": [ { "nodeId": 102, "nodeType": "SaveVideo", "file": "video_hunyuan_video_1.5_720p_t2v.mp4", "mediaType": "video" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0, "thumbnail": [ "output/video_hunyuan_video_1.5_720p_t2v.mp4" ] }, { "name": "wan2.1_fun_control", "title": "Wan 2.1 ControlNet", "description": "使用Wan 2.1 ControlNet通过姿态、深度和边缘控制引导生成视频。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "hoverDissolve", "tutorialUrl": "https://docs.comfy.org/tutorials/video/wan/fun-control", "tags": [ "视频转视频", "视频" ], "models": [ "Wan2.1", "Wan" ], "date": "2025-04-15", "size": 11381663334, "vram": 11381663334, "usage": 41, "io": { "inputs": [ { "nodeId": 52, "nodeType": "LoadImage", "file": "wan2.1_fun_control_start_frame.png", "mediaType": "image" }, { "nodeId": 80, "nodeType": "LoadVideo", "file": "wan2.1_fun_control_control_video.mp4", "mediaType": "video" } ] }, "username": "ComfyUI" }, { "name": "api_kling_dual_characters_effects", "title": "Kling：双重角色效果", "description": "并排上传两张角色图像。生成角色拥抱、亲吻或一起比心手势的视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "视频" ], "models": [ "Kling 1.6", "Kling" ], "logos": [ { "provider": "Kling" } ], "date": "2026-03-02", "openSource": false, "size": 0, "vram": 0, "usage": 1, "searchRank": 0, "io": { "inputs": [ { "nodeId": 13, "nodeType": "LoadImage", "file": "deep_curl_portrait.png", "mediaType": "image" }, { "nodeId": 14, "nodeType": "LoadImage", "file": "blue_tone_woman.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 17, "nodeType": "SaveVideo", "file": "api_kling_dual_characters_effects.mp4", "mediaType": "video" } ] }, "username": "ComfyUI", "thumbnail": [ "output/api_kling_dual_characters_effects.mp4" ] }, { "name": "video_wan_vace_14B_ref2v", "title": "Wan VACE参考到视频", "description": "创建与参考图像风格和内容匹配的视频。完美适用于风格一致的视频生成。", "mediaType": "image", "mediaSubtype": "webp", "tutorialUrl": "https://docs.comfy.org/tutorials/video/wan/vace", "tags": [ "视频", "图生视频" ], "models": [ "Wan2.1", "Wan" ], "date": "2025-05-21", "size": 57756572713, "vram": 57756572713, "usage": 17, "io": { "inputs": [ { "nodeId": 73, "nodeType": "LoadImage", "file": "video_wan_vace_14B_ref2v_reference_image.jpg", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "video_wan2_1_infinitetalk", "title": "InfiniteTalk：音频驱动的全身视频配音", "description": "上传源图像和目标音频。生成全身配音视频，同步动作，同时保留原始身份、背景和摄像机运动。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "音频转视频", "口型同步", "视频" ], "models": [ "Wan2.1 InfiniteTalk" ], "date": "2026-01-26", "openSource": true, "size": 29635274342, "vram": 29635274342, "usage": 48, "searchRank": 0, "io": { "inputs": [ { "nodeId": 24, "nodeType": "LoadAudio", "file": "audio_speaker1_woman.mp3", "mediaType": "audio" }, { "nodeId": 32, "nodeType": "LoadImage", "file": "two_character_talking.png", "mediaType": "image" }, { "nodeId": 90, "nodeType": "LoadAudio", "file": "audio_speaker2_man.mp3", "mediaType": "audio" } ] }, "username": "ComfyUI" }, { "name": "video_wanmove_480p", "title": "Wan-Move 运动控制图生视频", "description": "使用 Wan-Move 从单张图片生成视频，并通过轨迹引导实现精细的点级运动控制。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生视频", "运动控制", "视频" ], "models": [ "Wan", "Wan2.1" ], "date": "2025-12-15", "size": 25447681229, "vram": 25447681229, "usage": 14, "io": { "inputs": [ { "nodeId": 238, "nodeType": "LoadImage", "file": "snowboard_rush.png", "mediaType": "image" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0 }, { "name": "api_wan_r2v", "title": "WAN2.6: 角色参考视频生成", "description": "创建保留角色特点、动作自然且具电影感的视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "角色参考", "视频" ], "date": "2026-01-03", "openSource": false, "models": [ "Wan2.6", "Wan" ], "usage": 17, "io": { "inputs": [ { "nodeId": 34, "nodeType": "LoadVideo", "file": "armored_warrior.mp4", "mediaType": "video" }, { "nodeId": 35, "nodeType": "LoadVideo", "file": "spear_warrior.mp4", "mediaType": "video" } ] }, "username": "ComfyUI" }, { "name": "video_wan_vace_outpainting", "title": "Wan VACE外扩", "description": "使用Wan VACE外绘通过扩展视频大小生成扩展视频。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tutorialUrl": "https://docs.comfy.org/tutorials/video/wan/vace", "tags": [ "外绘", "视频" ], "models": [ "Wan2.1", "Wan" ], "date": "2025-05-21", "size": 57756572713, "vram": 57756572713, "usage": 33, "io": { "inputs": [ { "nodeId": 71, "nodeType": "LoadVideo", "file": "video_wan_vace_outpainting_input_video.mp4", "mediaType": "video" } ] }, "username": "ComfyUI" }, { "name": "video_wan2_2_5B_fun_control", "title": "Wan 2.2 5B Fun Control", "description": "多条件视频控制，支持姿态、深度和边缘引导。紧凑的5B体积，适用于实验开发。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生视频", "视频", "ControlNet" ], "models": [ "Wan2.2", "Wan" ], "date": "2025-07-29", "size": 18146236826, "vram": 18146236826, "usage": 82, "io": { "inputs": [ { "nodeId": 66, "nodeType": "LoadVideo", "file": "video_wan2_2_5B_fun_control_control_video.mp4", "mediaType": "video" }, { "nodeId": 70, "nodeType": "LoadImage", "file": "video_wan2_2_5B_fun_control_start_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_bytedance_seedance1_5_text_to_video", "title": "Seedance1.5 Pro：文生视频", "description": "通过文本创建完整的视听故事，包含同步语音、音乐和音效。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "文生视频" ], "models": [ "Seedance" ], "date": "2026-01-19", "openSource": false, "usage": 93, "username": "ComfyUI" }, { "name": "video_humo", "title": "HuMo 视频生成", "description": "基于音频、图像和文本生成视频，保持角色嘴唇同步。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "视频" ], "models": [ "HuMo" ], "date": "2025-09-21", "size": 27895812588, "vram": 27895812588, "usage": 9, "io": { "inputs": [ { "nodeId": 49, "nodeType": "LoadImage", "file": "video_humo_reference_image.png", "mediaType": "image" }, { "nodeId": 58, "nodeType": "LoadAudio", "file": "video_humo_input_audio.wav", "mediaType": "audio" } ] }, "username": "ComfyUI" }, { "name": "text_to_video_wan", "title": "Wan 2.1文生视频", "description": "使用Wan 2.1从文本提示生成视频。", "mediaType": "image", "mediaSubtype": "webp", "tutorialUrl": "https://docs.comfy.org/tutorials/video/wan/wan-video", "tags": [ "文生视频", "视频" ], "models": [ "Wan2.1", "Wan" ], "date": "2025-03-01", "size": 9824737690, "vram": 9824737690, "usage": 24, "username": "ComfyUI" }, { "name": "video_wan2_2_14B_fun_camera", "title": "Wan 2.2 14B Fun Camera控制", "description": "使用Wan 2.2 Fun Camera控制生成包含平移、缩放和旋转等摄像机运动控制的视频。", "mediaType": "image", "mediaSubtype": "webp", "tutorialUrl": "https://docs.comfy.org/tutorials/video/wan/wan2-2-fun-camera", "tags": [ "视频转视频", "视频" ], "models": [ "Wan2.2", "Wan" ], "date": "2025-08-17", "size": 40050570035, "vram": 40050570035, "usage": 15, "io": { "inputs": [ { "nodeId": 79, "nodeType": "LoadImage", "file": "video_wan2_2_14B_fun_camera_start_image.jpg", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_vidu_reference_to_video", "title": "Vidu Q1：参考生视频", "description": "使用多个参考图像（最多7张）生成具有一致主题的视频，确保角色和风格在整个视频序列中的连续性。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "视频", "图生视频", "API" ], "models": [ "Vidu Q1", "Vidu" ], "date": "2025-08-23", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 1, "io": { "inputs": [ { "nodeId": 28, "nodeType": "LoadImage", "file": "api_vidu_reference_to_video_reference_image_2.jpg", "mediaType": "image" }, { "nodeId": 30, "nodeType": "LoadImage", "file": "api_vidu_reference_to_video_reference_image_1.jpg", "mediaType": "image" }, { "nodeId": 45, "nodeType": "LoadImage", "file": "api_vidu_reference_to_video_reference_image_3.jpg", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "image_to_video_wan", "title": "Wan 2.1图生视频", "description": "使用Wan 2.1从图像生成视频。", "mediaType": "image", "mediaSubtype": "webp", "tutorialUrl": "https://docs.comfy.org/tutorials/video/wan/wan-video", "tags": [ "文生视频", "视频" ], "models": [ "Wan2.1", "Wan" ], "date": "2025-03-01", "size": 41049149932, "vram": 41049149932, "usage": 46, "io": { "inputs": [ { "nodeId": 52, "nodeType": "LoadImage", "file": "image_to_video_wan_start_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_kling2_6_t2v", "title": "Kling2.6：文生视频（音画同步）", "description": "将你的故事转化为含对话、音乐、音效和环境音的视频", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "音频", "文生视频", "视频" ], "models": [ "Kling", "Kling 2.6" ], "date": "2025-12-22", "openSource": false, "size": 0, "vram": 0, "usage": 9, "searchRank": 0, "username": "ComfyUI" }, { "name": "api_runway_first_last_frame", "title": "Runway：首尾帧到视频", "description": "使用Runway的精确度在两个关键帧之间生成流畅的视频过渡。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "视频", "API", "首尾帧视频" ], "models": [ "Runway" ], "date": "2025-03-01", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 27, "io": { "inputs": [ { "nodeId": 10, "nodeType": "LoadImage", "file": "api_runway_first_last_frame_first_frame.jpg", "mediaType": "image" }, { "nodeId": 11, "nodeType": "LoadImage", "file": "api_runway_first_last_frame_last_frame.jpg", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_runway_gen3a_turbo_image_to_video", "title": "Runway：Gen3a Turbo图生视频", "description": "使用Runway Gen3a Turbo从静态图像生成电影级视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生视频", "视频", "API" ], "models": [ "Runway" ], "date": "2025-03-01", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 7, "io": { "inputs": [ { "nodeId": 4, "nodeType": "LoadImage", "file": "api_runway_gen3a_turbo_image_to_video_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "wan2.1_flf2v_720_f16", "title": "Wan 2.1 首尾帧视频 720p F16", "description": "使用Wan 2.1 FLF2V通过控制首尾帧生成视频。", "mediaType": "image", "mediaSubtype": "webp", "tutorialUrl": "https://docs.comfy.org/tutorials/video/wan/wan-flf", "tags": [ "首尾帧视频", "视频" ], "models": [ "Wan2.1", "Wan" ], "date": "2025-04-15", "size": 41049149932, "vram": 41049149932, "usage": 21, "io": { "inputs": [ { "nodeId": 52, "nodeType": "LoadImage", "file": "wan2.1_flf2v_720_f16_start_image.png", "mediaType": "image" }, { "nodeId": 72, "nodeType": "LoadImage", "file": "wan2.1_flf2v_720_f16_end_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "video_kandinsky5_i2v", "title": "Kandinsky 5.0 Video Lite 图生视频", "description": "一款轻量级2B模型，可根据英文和俄文提示生成高质量视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生视频", "视频" ], "models": [ "Kandinsky" ], "date": "2025-12-09", "size": 14710262989, "vram": 14710262989, "usage": 18, "searchRank": 0, "openSource": true, "io": { "inputs": [ { "nodeId": 11, "nodeType": "LoadImage", "file": "crystal_flower.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_vidu_q2_t2v", "title": "Vidu Q2：文生视频", "description": "使用Vidu Q2模型从文本提示生成高质量1080p视频", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "文生视频", "视频" ], "models": [ "Vidu Q2", "Vidu" ], "date": "2026-01-12", "openSource": false, "usage": 2, "username": "ComfyUI" }, { "name": "api_veo2_i2v", "title": "Veo2：图生视频", "description": "使用Google Veo2 API从图像生成视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生视频", "视频", "API" ], "models": [ "Veo", "Google" ], "date": "2025-03-01", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 21, "io": { "inputs": [ { "nodeId": 61, "nodeType": "LoadImage", "file": "api_veo2_i2v_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_wan_text_to_video", "title": "Wan2.5: 文生视频", "description": "生成具有同步音频、增强运动和卓越质量的视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生视频", "视频", "API" ], "models": [ "Wan2.5", "Wan" ], "date": "2025-09-27", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 4, "io": { "inputs": [ { "nodeId": 3, "nodeType": "LoadAudio", "file": "audio_ace_step_1_m2m_editing_input_audio.mp3", "mediaType": "audio" } ] }, "username": "ComfyUI" }, { "name": "api_vidu_start_end_to_video", "title": "Vidu Q1：首尾帧视频", "description": "在定义的起始和结束帧之间创建流畅的视频过渡，具有自然运动插值和一致的视觉质量。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "视频", "API", "首尾帧视频" ], "models": [ "Vidu Q1", "Vidu" ], "date": "2025-08-23", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 5, "io": { "inputs": [ { "nodeId": 28, "nodeType": "LoadImage", "file": "api_vidu_start_end_to_video_start_frame.jpg", "mediaType": "image" }, { "nodeId": 30, "nodeType": "LoadImage", "file": "api_vidu_start_end_to_video_last_frame.jpg", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "video_wan_vace_flf2v", "title": "Wan VACE首尾帧", "description": "通过定义开始和结束帧生成平滑的视频过渡。支持自定义关键帧序列。", "mediaType": "image", "mediaSubtype": "webp", "tutorialUrl": "https://docs.comfy.org/tutorials/video/wan/vace", "tags": [ "首尾帧视频", "视频" ], "models": [ "Wan2.1", "Wan" ], "date": "2025-05-21", "size": 57756572713, "vram": 57756572713, "usage": 29, "io": { "inputs": [ { "nodeId": 144, "nodeType": "LoadImage", "file": "video_wan_vace_flf2v_start_image.png", "mediaType": "image" }, { "nodeId": 147, "nodeType": "LoadImage", "file": "video_wan_vace_flf2v_end_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "video_kandinsky5_t2v", "title": "Kandinsky 5.0 Video Lite 文生视频", "description": "一款轻量级2B模型，可根据英文和俄文提示生成高质量视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生视频", "视频" ], "models": [ "Kandinsky" ], "date": "2025-12-09", "size": 14710262989, "vram": 14710262989, "usage": 3, "searchRank": 0, "openSource": true, "username": "ComfyUI" }, { "name": "video_capybara_v0_1_video_edit", "title": "Capybara：视频编辑", "description": "上传图像或视频及文本指令以执行生成或编辑任务。根据提供的指令生成新的或编辑后的图像或视频。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "视频编辑" ], "models": [ "Capybara" ], "date": "2026-02-23", "openSource": true, "size": 37044092928, "vram": 37044092928, "usage": 27, "searchRank": 0, "io": { "inputs": [ { "nodeId": 127, "nodeType": "LoadVideo", "file": "horse_running.mp4", "mediaType": "video" } ], "outputs": [ { "nodeId": 114, "nodeType": "SaveVideo", "file": "video_capybara_v0_1_video_edit.mp4", "mediaType": "video" } ] }, "username": "ComfyUI", "thumbnail": [ "input/horse_running.mp4", "output/video_capybara_v0_1_video_edit.mp4" ] }, { "name": "video_capybara_v0_1_image_to_video", "title": "Capybara：图生视频", "description": "上传一张图片并提供文字指令。生成一个新视频，其中输入内容会根据您的指令进行编辑，同时保持结构和时间连贯性。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "视频", "图生视频" ], "models": [ "Capybara" ], "date": "2026-02-23", "openSource": true, "size": 37044092928, "vram": 37044092928, "usage": 42, "searchRank": 0, "io": { "inputs": [ { "nodeId": 126, "nodeType": "LoadImage", "file": "joy_snap.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 114, "nodeType": "SaveVideo", "file": "video_capybara_v0_1_image_to_video.mp4", "mediaType": "video" } ] }, "username": "ComfyUI", "thumbnail": [ "output/video_capybara_v0_1_image_to_video.mp4" ] }, { "name": "api_pixverse_t2v", "title": "PixVerse：文生视频", "description": "生成具有准确提示解释和惊人视频动态的视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生视频", "视频", "API" ], "models": [ "PixVerse" ], "date": "2025-03-01", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 18, "username": "ComfyUI" }, { "name": "video_wan2.1_alpha_t2v_14B", "title": "Wan2.1 Alpha 文生视频", "description": "使用 alpha 通道支持透明背景和半透明对象的文生视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生视频", "视频" ], "models": [ "Wan2.1", "Wan" ], "date": "2025-10-06", "size": 22494891213, "vram": 22494891213, "usage": 1, "username": "ComfyUI" }, { "name": "txt_to_image_to_video", "title": "SVD文生图到视频", "mediaType": "image", "mediaSubtype": "webp", "description": "通过首先从文本提示创建图像来生成视频。", "tutorialUrl": "https://comfyanonymous.github.io/ComfyUI_examples/video/#image-to-video", "tags": [ "文生视频", "视频" ], "models": [ "SVD", "Stability" ], "date": "2025-03-01", "size": 16492674417, "vram": 16492674417, "usage": 1, "username": "ComfyUI" }, { "name": "api_hailuo_minimax_i2v", "title": "MiniMax：图生视频", "description": "使用MiniMax通过CGI集成从图像和文本生成精美的视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生视频", "视频", "API" ], "models": [ "MiniMax" ], "date": "2025-03-01", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 8, "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadImage", "file": "api_hailuo_minimax_i2v_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_luma_i2v", "title": "Luma：图生视频", "description": "获取静态图像并立即创建神奇的高质量动画。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生视频", "视频", "API" ], "models": [ "Luma" ], "date": "2025-03-01", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 19, "io": { "inputs": [ { "nodeId": 9, "nodeType": "LoadImage", "file": "api_luma_i2v_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "video_wan_ati", "title": "Wan ATI", "description": "轨迹控制视频生成。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "hoverDissolve", "tutorialUrl": "https://docs.comfy.org/tutorials/video/wan/wan-ati", "tags": [ "视频" ], "models": [ "Wan2.1", "Wan" ], "date": "2025-05-21", "size": 25393994138, "vram": 25393994138, "usage": 3, "io": { "inputs": [ { "nodeId": 240, "nodeType": "LoadImage", "file": "video_wan_ati_input_image.jpg", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_ltxv_text_to_video", "title": "LTX-2：文生视频", "description": "使用Lightricks LTX-2和同步音频，根据文本提示生成高质量视频。支持最高4K分辨率和50帧/秒，满足快速、专业和超高清等多种制作需求。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生视频", "视频", "API" ], "models": [ "LTX-2", "Lightricks" ], "date": "2025-10-28", "openSource": false, "size": 0, "vram": 0, "usage": 5, "username": "ComfyUI" }, { "name": "api_vidu_text_to_video", "title": "Vidu Q1：文生视频", "description": "使用Vidu的高级AI模型，通过可调节的运动幅度和持续时间控制，从文本提示生成高质量1080p视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生视频", "视频", "API" ], "models": [ "Vidu Q1", "Vidu" ], "date": "2025-08-23", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 8, "username": "ComfyUI" }, { "name": "api_bytedance_text_to_video", "title": "ByteDance: 文生视频", "description": "使用 ByteDance 的 Seedance 模型直接从文本提示生成高质量视频。支持多种分辨率和宽高比，具有自然运动和电影级质量。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "视频", "API", "文生视频" ], "models": [ "Seedance1.0 Pro", "ByteDance" ], "date": "2025-10-06", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 7, "username": "ComfyUI" }, { "name": "api_vidu_image_to_video", "title": "Vidu Q1：图生视频", "description": "使用Vidu将静态图像转换为具有精确运动控制和可定制运动幅度的动态1080p视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生视频", "视频", "API" ], "models": [ "Vidu Q1", "Vidu" ], "date": "2025-08-23", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 3, "io": { "inputs": [ { "nodeId": 22, "nodeType": "LoadImage", "file": "api_vidu_image_to_video_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_hailuo_minimax_t2v", "title": "MiniMax：文生视频", "description": "直接从文本提示生成高质量视频。探索MiniMax的高级AI功能，利用专业CGI效果和风格元素创建多样化的视觉叙事，让您的描述生动起来。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生视频", "视频", "API" ], "models": [ "MiniMax" ], "date": "2025-03-01", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 1, "username": "ComfyUI" }, { "name": "video_wan2_2_5B_fun_inpaint", "title": "Wan 2.2 5B Fun Inpaint", "description": "从开始和结束帧高效地进行视频修复。5B 模型提供快速迭代以测试工作流程。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生视频", "视频" ], "models": [ "Wan2.2", "Wan" ], "date": "2025-07-29", "size": 18146236826, "vram": 18146236826, "usage": 2, "io": { "inputs": [ { "nodeId": 70, "nodeType": "LoadImage", "file": "video_wan2_2_5B_fun_inpaint_start_image.png", "mediaType": "image" }, { "nodeId": 74, "nodeType": "LoadImage", "file": "video_wan2_2_5B_fun_inpaint_end_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_kling_effects", "title": "Kling：视频效果", "description": "使用Kling通过对图像应用视觉效果生成动态视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "视频", "API" ], "models": [ "Kling", "Kling 1.6" ], "date": "2025-03-01", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 5, "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadImage", "file": "api_kling_effects_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_hailuo_minimax_video", "title": "MiniMax：视频", "description": "使用MiniMax Hailuo-02模型，通过可选的首帧控制从文本提示生成高质量视频。支持多种分辨率（768P/1080P）和时长（6/10秒），具有智能提示优化功能。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "文生视频", "视频" ], "models": [ "MiniMax" ], "date": "2025-03-01", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 14, "io": { "inputs": [ { "nodeId": 13, "nodeType": "LoadImage", "file": "api_hailuo_minimax_video_first_frame.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 14, "nodeType": "SaveVideo", "file": "api_hailuo_minimax_video.mp4", "mediaType": "video" } ] }, "username": "ComfyUI", "logos": [ { "provider": "MiniMax" } ], "searchRank": 0, "thumbnail": [ "output/api_hailuo_minimax_video.mp4" ] }, { "name": "ltxv_text_to_video", "title": "LTXV文生视频", "mediaType": "image", "mediaSubtype": "webp", "description": "从文本提示生成视频。", "tutorialUrl": "https://docs.comfy.org/tutorials/video/ltxv", "tags": [ "文生视频", "视频" ], "models": [ "LTX-0.9.5" ], "date": "2025-03-01", "size": 19155554140, "vram": 19155554140, "usage": 54, "username": "ComfyUI" }, { "name": "api_luma_t2v", "title": "Luma：文生视频", "description": "可以使用简单的提示生成高质量视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生视频", "视频", "API" ], "models": [ "Luma" ], "date": "2025-03-01", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 3, "username": "ComfyUI" }, { "name": "api_pixverse_i2v", "title": "PixVerse：图生视频", "description": "使用PixVerse从静态图像生成具有运动和效果的动态视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生视频", "视频", "API" ], "models": [ "PixVerse" ], "date": "2025-03-01", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 2, "io": { "inputs": [ { "nodeId": 6, "nodeType": "LoadImage", "file": "api_runway_reference_to_image_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_pixverse_template_i2v", "title": "PixVerse模板：图生视频", "description": "使用PixVerse从静态图像生成具有运动和效果的动态视频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生视频", "视频", "API" ], "models": [ "PixVerse" ], "date": "2025-03-01", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 2, "io": { "inputs": [ { "nodeId": 34, "nodeType": "LoadImage", "file": "api_pixverse_template_i2v_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "video_wan2.1_fun_camera_v1.1_14B", "title": "Wan 2.1 Fun Camera 14B", "description": "使用完整的14B模型生成具有高级摄像机控制的高质量视频", "mediaType": "image", "mediaSubtype": "webp", "tutorialUrl": "https://docs.comfy.org/tutorials/video/wan/fun-control", "tags": [ "视频" ], "models": [ "Wan2.1", "Wan" ], "date": "2025-04-15", "size": 42047729828, "vram": 42047729828, "usage": 4, "io": { "inputs": [ { "nodeId": 52, "nodeType": "LoadImage", "file": "video_wan2.1_fun_camera_v1.1_14B_start_image.jpg", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "video_wan_vace_14B_t2v", "title": "Wan VACE文生视频", "description": "将文本描述转换为高质量视频。VACE-14B模型支持480p和720p两种分辨率。", "mediaType": "image", "mediaSubtype": "webp", "tutorialUrl": "https://docs.comfy.org/tutorials/video/wan/vace", "tags": [ "文生视频", "视频" ], "models": [ "Wan2.1", "Wan" ], "date": "2025-05-21", "size": 57756572713, "vram": 57756572713, "usage": 4, "username": "ComfyUI" }, { "name": "video_wan2.1_fun_camera_v1.1_1.3B", "title": "Wan 2.1 Fun Camera 1.3B", "description": "使用Wan 2.1 Fun Camera 1.3B模型生成具有电影级摄像机运动的动态视频。", "mediaType": "image", "mediaSubtype": "webp", "tutorialUrl": "https://docs.comfy.org/tutorials/video/wan/fun-control", "tags": [ "视频" ], "models": [ "Wan2.1", "Wan" ], "date": "2025-04-15", "size": 11489037517, "vram": 11489037517, "usage": 3, "io": { "inputs": [ { "nodeId": 52, "nodeType": "LoadImage", "file": "video_wan2.1_fun_camera_v1.1_1.3B_start_image.jpg", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "wan2.1_fun_inp", "title": "Wan 2.1修复", "description": "使用Wan 2.1修复从开始和结束帧生成视频。", "mediaType": "image", "mediaSubtype": "webp", "tutorialUrl": "https://docs.comfy.org/tutorials/video/wan/fun-inp", "tags": [ "图像修复", "视频" ], "models": [ "Wan2.1", "Wan" ], "date": "2025-04-15", "size": 11381663334, "vram": 11381663334, "usage": 3, "io": { "inputs": [ { "nodeId": 52, "nodeType": "LoadImage", "file": "wan2.1_fun_inp_start_image.png", "mediaType": "image" }, { "nodeId": 72, "nodeType": "LoadImage", "file": "wan2.1_fun_inp_end_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "hunyuan_video_text_to_video", "title": "Hunyuan视频文生视频", "mediaType": "image", "mediaSubtype": "webp", "description": "使用Hunyuan模型从文本提示生成视频。", "tutorialUrl": "https://comfyanonymous.github.io/ComfyUI_examples/hunyuan_video/", "tags": [ "文生视频", "视频" ], "models": [ "Hunyuan Video", "Tencent" ], "date": "2025-03-01", "size": 35476429865, "vram": 35476429865, "usage": 4, "username": "ComfyUI" } ] }, { "moduleName": "default", "type": "audio", "category": "生成类型", "icon": "icon-[lucide--volume-2]", "title": "音频", "templates": [ { "name": "audio_stable_audio_3_medium_base", "title": "Stable Audio 3.0 Medium Base", "description": "输入声音、音乐或效果的简短文本描述。该工作流使用Qwen扩展您的提示，并从Stable Audio 3生成立体声音频片段。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文本转音频" ], "models": [ "Stable Audio" ], "date": "2026-05-20", "openSource": true, "size": 15891378995, "vram": 15891378995, "usage": 215, "searchRank": 0, "username": "ComfyUI" }, { "name": "audio_stable_audio_3_medium", "title": "Stable Audio 3.0 Medium", "description": "输入一个简短的文本创意、可选时长、种子和类别。使用 Stable Audio 3 生成立体声音频（音乐、音效或乐器），并可选 AI 驱动的文本扩展。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文本转音频" ], "models": [ "Stable Audio" ], "date": "2026-05-20", "openSource": true, "size": 15676630630, "vram": 15676630630, "usage": 131, "searchRank": 0, "username": "ComfyUI" }, { "name": "api_sonilo_v2m", "title": "Sonilo：视频配乐生成", "description": "从视频素材生成同步配乐。输入视频以制作与其节奏和情感线索相匹配的音乐，输出完美定时的音频文件。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "音频", "音乐", "原声带" ], "models": [ "Sonilo" ], "logos": [ { "provider": "Sonilo" } ], "date": "2026-04-12", "openSource": false, "size": 0, "vram": 0, "usage": 39, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 705, "nodeType": "LoadVideo", "file": "leopard_rider.mp4", "mediaType": "video" } ] } }, { "name": "api_sonilo_t2m", "title": "Sonilo：文生音乐", "description": "通过文本提示生成高质量、可用于制作的高品质音乐。输入描述性文本，创建具有流媒体播放和精确时长控制的原创配乐。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "音频", "原声带", "文生音乐" ], "models": [ "Sonilo" ], "logos": [ { "provider": "Sonilo" } ], "date": "2026-04-12", "openSource": false, "size": 0, "vram": 0, "usage": 27, "searchRank": 0, "username": "ComfyUI" }, { "name": "audio_ace_step1_5_xl_base", "title": "ACE-Step 1.5XL 基础版：文本转音乐", "description": "使用40亿参数ACE-Step 1.5 XL模型，从文本提示生成高质量音频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生音乐" ], "models": [ "ACE-Step" ], "date": "2026-04-10", "openSource": true, "size": 19864223744, "vram": 19864223744, "usage": 1, "searchRank": 0, "username": "ComfyUI", "includeOnDistributions": [ "local" ] }, { "name": "audio_ace_step1_5_xl_turbo", "title": "ACE-Step 1.5XL Turbo：文本转音乐", "description": "使用蒸馏的4B ACE-Step模型，从文本提示生成高质量音乐。仅需8次推理步骤即可生成商业级音频，无需CFG。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "音乐", "文生音乐" ], "models": [ "ACE-Step" ], "date": "2026-04-10", "openSource": true, "size": 19864223744, "vram": 19864223744, "usage": 197, "searchRank": 0, "username": "ComfyUI" }, { "name": "audio_ace_step1_5_xl_sft", "title": "ACE-Step 1.5XL SFT：文生音乐", "description": "使用40亿参数SFT模型，从文本提示生成高质量音乐。通过CFG控制输出，确保精确遵循提示。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生音乐" ], "models": [ "ACE-Step" ], "date": "2026-04-10", "openSource": true, "size": 19864223744, "vram": 19864223744, "usage": 65, "searchRank": 0, "username": "ComfyUI" }, { "name": "api_elevenlabs_text_to_speech", "title": "ElevenLabs：文生语音", "description": "输入文本以生成超逼真语音，或上传语音样本以克隆用于合成。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文本转语音", "音频", "API" ], "models": [ "ElevenLabs" ], "logos": [ { "provider": "ElevenLabs" } ], "date": "2026-02-23", "openSource": false, "size": 0, "vram": 0, "usage": 272, "searchRank": 0, "io": { "inputs": [ { "nodeId": 210, "nodeType": "LoadAudio", "file": "voice_demo.mp3", "mediaType": "audio" } ] }, "username": "ComfyUI", "thumbnail": [ "thumbnail/api_elevenlabs_text_to_speech.png" ] }, { "name": "api_elevenlabs_text_to_dialogue", "title": "ElevenLabs：文本转对话", "description": "输入文本并选择语音配置文件，生成高质量、情感丰富的音频对话。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "文本转语音" ], "models": [ "ElevenLabs" ], "logos": [ { "provider": "ElevenLabs" } ], "date": "2026-02-23", "openSource": false, "size": 0, "vram": 0, "usage": 90, "searchRank": 0, "io": { "inputs": [ { "nodeId": 210, "nodeType": "LoadAudio", "file": "voice_demo.mp3", "mediaType": "audio" } ] }, "username": "ComfyUI", "thumbnail": [ "thumbnail/api_elevenlabs_text_to_dialogue.png" ] }, { "name": "api_elevenlabs_speech_to_speech", "title": "ElevenLabs：语音转语音", "description": "上传源音频文件以应用新的语音语调或克隆不同的声音，生成修改后的音频输出。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "语音克隆", "API" ], "models": [ "ElevenLabs" ], "logos": [ { "provider": "ElevenLabs" } ], "date": "2026-02-23", "openSource": false, "size": 0, "vram": 0, "usage": 37, "searchRank": 0, "io": { "inputs": [ { "nodeId": 210, "nodeType": "LoadAudio", "file": "voice_demo.mp3", "mediaType": "audio" }, { "nodeId": 221, "nodeType": "LoadAudio", "file": "api_elevenlabs_speech_to_speech.mp3", "mediaType": "audio" } ] }, "username": "ComfyUI", "thumbnail": [ "thumbnail/api_elevenlabs_speech_to_speech.png" ] }, { "name": "api_elevenlabs_voice_isolation", "title": "ElevenLabs：语音隔离", "description": "上传包含背景噪音的音频文件。使用 ElevenLabs API 分离并输出干净的语音轨道。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "语音隔离" ], "models": [ "ElevenLabs" ], "logos": [ { "provider": "ElevenLabs" } ], "date": "2026-02-21", "openSource": false, "size": 0, "vram": 0, "usage": 7, "searchRank": 0, "io": { "inputs": [ { "nodeId": 131, "nodeType": "LoadAudio", "file": "noisy_background.mp3", "mediaType": "audio" } ] }, "username": "ComfyUI", "thumbnail": [ "thumbnail/api_elevenlabs_voice_isolation-1.png", "thumbnail/api_elevenlabs_voice_isolation-2.png" ] }, { "name": "api_elevenLabs_speech_to_text", "title": "ElevenLabs：语音转文本", "description": "上传音频或视频文件，将其语音转录为准确、可编辑的文本。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "语音转文本" ], "models": [ "ElevenLabs" ], "logos": [ { "provider": "ElevenLabs" } ], "date": "2026-02-23", "openSource": false, "size": 0, "vram": 0, "usage": 119, "searchRank": 0, "io": { "inputs": [ { "nodeId": 219, "nodeType": "LoadAudio", "file": "voice_demo.mp3", "mediaType": "audio" } ] }, "username": "ComfyUI", "thumbnail": [ "thumbnail/api_elevenLabs_speech_to_text.png" ] }, { "name": "api_elevenlabs_text_to_sound_effects", "title": "ElevenLabs：文生音效", "description": "输入文本提示以生成自定义音效和环境音频。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "音频", "音效", "API" ], "models": [ "ElevenLabs" ], "logos": [ { "provider": "ElevenLabs" } ], "date": "2026-02-21", "openSource": false, "size": 0, "vram": 0, "usage": 39, "searchRank": 0, "io": { "outputs": [ { "nodeId": 137, "nodeType": "SaveAudioMP3", "file": "api_elevenlabs_text_to_sound_effects.mp3", "mediaType": "audio" } ] }, "username": "ComfyUI", "thumbnail": [ "thumbnail/api_elevenlabs_text_to_sound_effects.png" ] }, { "name": "audio_ace_step_1_5_split_llm", "title": "ACE-Step 1.5 AI 歌词歌曲生成", "description": "输入音乐风格和可选歌词。使用AI生成完整歌曲，包括歌词和音频。", "mediaType": "audio", "mediaSubtype": "mp3", "tags": [ "API", "文本转音频" ], "models": [ "ACE-Step", "Google Gemini" ], "logos": [ { "provider": "Google" } ], "date": "2026-02-09", "openSource": false, "size": 9985798963, "vram": 9985798963, "usage": 21, "searchRank": 0, "io": { "outputs": [ { "nodeId": 1, "nodeType": "SaveAudioMP3", "file": "audio_ace_step_1_5_split_llm.mp3", "mediaType": "audio" } ] }, "username": "ComfyUI" }, { "name": "audio_ace_step_1_5_split_4b", "title": "ACE-Step 1.5 音乐生成（4B LLM）", "description": "输入文本提示以生成音乐。与较小版本相比，此4B模型版本具备更强的音频理解和作曲能力。", "mediaType": "audio", "mediaSubtype": "mp3", "tags": [ "音频", "音乐", "文本转音频" ], "models": [ "ACE-Step" ], "date": "2026-02-05", "openSource": true, "size": 14710262989, "vram": 14710262989, "usage": 53, "searchRank": 0, "username": "ComfyUI" }, { "name": "audio_ace_step_1_5_split", "title": "ACE-Step 1.5 音乐生成工作流程", "description": "输入描述音乐风格和可选歌词的文本提示。在消费级硬件上10秒内生成完整的高质量音频歌曲。", "mediaType": "audio", "mediaSubtype": "mp3", "tags": [ "音频", "文本转音频", "音乐" ], "models": [ "ACE-Step" ], "date": "2026-02-02", "openSource": true, "size": 10737418240, "vram": 2147483648, "usage": 36, "searchRank": 0, "username": "ComfyUI" }, { "name": "audio_ace_step_1_5_checkpoint", "title": "ACE-Step 1.5 音乐生成 AIO", "description": "输入风格标签和歌词以生成完整歌曲。该工作流程使用ACE-Step 1.5模型，在消费级硬件上10秒内生成商业级音乐。", "mediaType": "audio", "mediaSubtype": "mp3", "tags": [ "音频", "音乐", "文本转音频" ], "models": [ "ACE-Step" ], "date": "2026-02-03", "openSource": true, "size": 10737418240, "vram": 5368709120, "usage": 18, "searchRank": 0, "username": "ComfyUI" }, { "name": "api_stability_ai_audio_to_audio", "title": "Stable Audio 2.5：音频转音频", "description": "使用 Stable Audio 2.5 将现有音频转化为全新作品，上传音频，AI 自动生成完整音轨。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "音频", "音频转音频" ], "date": "2025-09-09", "models": [ "Stability", "Stable Audio" ], "openSource": false, "size": 0, "vram": 0, "usage": 52, "io": { "inputs": [ { "nodeId": 11, "nodeType": "LoadAudio", "file": "api_stability_ai_audio_to_audio_input_audio.mp3", "mediaType": "audio" } ] }, "searchRank": 0, "username": "ComfyUI" }, { "name": "api_stability_ai_text_to_audio", "title": "Stable Audio 2.5：文生音频", "description": "使用 Stable Audio 2.5 从文本描述生成音乐，几秒钟内创作数分钟的音轨。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "音频", "文本转音频" ], "date": "2025-09-09", "models": [ "Stability", "Stable Audio" ], "openSource": false, "size": 0, "vram": 0, "usage": 12, "logos": [ { "provider": "Stability" } ], "searchRank": 0, "username": "ComfyUI" }, { "name": "audio_stable_audio_example", "title": "Stable Audio 1.0：文生音频", "mediaType": "image", "mediaSubtype": "webp", "description": "使用Stable Audio从文本提示生成音频。", "tags": [ "音频", "文本转音频" ], "models": [ "Stability", "Stable Audio" ], "date": "2025-03-01", "tutorialUrl": "https://comfyanonymous.github.io/ComfyUI_examples/audio/", "size": 5690831667, "vram": 5690831667, "usage": 20, "openSource": true, "searchRank": 0, "username": "ComfyUI" }, { "name": "audio_ace_step_1_m2m_editing", "title": "ACE Step v1 M2M编辑", "mediaType": "audio", "mediaSubtype": "mp3", "description": "使用ACE-Step v1 M2M编辑现有歌曲以改变风格和歌词。", "tags": [ "音频编辑", "音频" ], "models": [ "ACE-Step" ], "date": "2025-03-01", "tutorialUrl": "https://docs.comfy.org/tutorials/audio/ace-step/ace-step-v1", "size": 7698728878, "vram": 7698728878, "usage": 89, "io": { "inputs": [ { "nodeId": 64, "nodeType": "LoadAudio", "file": "audio_ace_step_1_m2m_editing_input_audio.mp3", "mediaType": "audio" } ] }, "username": "ComfyUI" }, { "name": "audio_ace_step_1_t2a_song", "title": "ACE Step v1文本到歌曲", "mediaType": "audio", "mediaSubtype": "mp3", "description": "使用ACE-Step v1从文本提示生成带有人声的歌曲，支持多语言和风格定制。", "tags": [ "文本转音频", "音频" ], "models": [ "ACE-Step" ], "date": "2025-03-01", "tutorialUrl": "https://docs.comfy.org/tutorials/audio/ace-step/ace-step-v1", "size": 7698728878, "vram": 7698728878, "usage": 5, "username": "ComfyUI" }, { "name": "audio_ace_step_1_t2a_instrumentals", "title": "ACE-Step v1文本到器乐音乐", "mediaType": "audio", "mediaSubtype": "mp3", "description": "使用ACE-Step v1从文本提示生成器乐音乐。", "tags": [ "文本转音频", "音频" ], "models": [ "ACE-Step" ], "date": "2025-03-01", "tutorialUrl": "https://docs.comfy.org/tutorials/audio/ace-step/ace-step-v1", "size": 7698728878, "vram": 7698728878, "usage": 5, "username": "ComfyUI" }, { "name": "api_stability_ai_audio_inpaint", "title": "Stability Audio 2.5：音频修复", "description": "使用 Stable Audio 2.5 扩展音频，上传音频，AI 自动补充扩展剩余部分。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "音频", "音频转音频" ], "date": "2025-09-09", "models": [ "Stability", "Stable Audio" ], "openSource": false, "size": 0, "vram": 0, "usage": 5, "io": { "inputs": [ { "nodeId": 11, "nodeType": "LoadAudio", "file": "api_stability_ai_audio_inpaint_input_audio.wav", "mediaType": "audio" } ] }, "logos": [ { "provider": "Stability" } ], "searchRank": 0, "username": "ComfyUI" } ] }, { "moduleName": "default", "type": "3d", "category": "生成类型", "icon": "icon-[lucide--box]", "title": "3D模型", "templates": [ { "name": "api_hunyuan3d_image_to_model", "title": "混元3D：图生模型", "description": "上传图像以生成带有几何和PBR纹理的3D模型。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "3D模型", "API", "图生模型" ], "models": [ "Hunyuan3D" ], "date": "2026-05-21", "openSource": false, "usage": 223, "io": { "inputs": [ { "nodeId": 1, "nodeType": "LoadImage", "file": "pink_robot_front.png", "mediaType": "image" }, { "nodeId": 10, "nodeType": "LoadImage", "file": "pink_robot_back.png", "mediaType": "image" } ] }, "username": "ComfyUI", "size": 0, "vram": 0, "searchRank": 0 }, { "name": "api_tripo3_1_image_to_model", "title": "Tripo H3.1：图生模型", "description": "上传物体的参考图像。生成具有密集几何结构和PBR材质的高细节3D模型。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "3D模型", "API" ], "models": [ "Tripo" ], "logos": [ { "provider": "Tripo" } ], "date": "2026-05-09", "openSource": false, "size": 0, "vram": 0, "usage": 179, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 10, "nodeType": "LoadImage", "file": "cybernetic_robot.png", "mediaType": "image" } ] } }, { "name": "api_tripo_p1_image_to_model", "title": "Tripo P1：图像到模型", "description": "上传您主体的参考图像。生成具有干净拓扑和优化多边形预算的、可用于游戏的3D模型。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "3D模型", "API", "图生3D模型" ], "models": [ "Tripo P1" ], "logos": [ { "provider": "Tripo" } ], "date": "2026-05-28", "openSource": false, "size": 0, "vram": 0, "usage": 0, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 45, "nodeType": "LoadImage", "file": "knight_with_sword.png", "mediaType": "image" } ] } }, { "name": "api_tripo_p1_text_to_model", "title": "Tripo P1：文本转模型", "description": "输入3D模型的文本描述。生成一个具有干净拓扑结构和优化多边形数量的游戏就绪3D模型。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "3D模型" ], "models": [ "Tripo P1" ], "logos": [ { "provider": "Tripo" } ], "date": "2026-05-28", "openSource": false, "size": 0, "vram": 0, "usage": 0, "searchRank": 0, "username": "ComfyUI" }, { "name": "api_rodin3d_gen2_5_image_to_3d", "title": "Rodin Gen2.5: Image to Model", "description": "Upload single or multi-view images to create a 3D model with adjustable quality and enhanced texture", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "3D模型", "API" ], "models": [ "Rodin" ], "logos": [ { "provider": "Rodin" } ], "date": "2026-05-23", "openSource": false, "size": 0, "vram": 0, "usage": 80, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadImage", "file": "bull_robot.png", "mediaType": "image" } ] } }, { "name": "api_tripo_p1_mv_to_model", "title": "Tripo P1：多视图到模型", "description": "上传一张图片或多视角参考图，生成干净的低多边形3D模型，具有可控的多边形数量和有序拓扑结构，可直接导出到游戏引擎。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "3D模型", "API", "图生3D模型" ], "models": [ "Tripo P1" ], "logos": [ { "provider": "Tripo" } ], "date": "2026-05-28", "openSource": false, "size": 0, "vram": 0, "usage": 0, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 45, "nodeType": "LoadImage", "file": "wooden_doll.png", "mediaType": "image" }, { "nodeId": 47, "nodeType": "LoadImage", "file": "wooden_doll_left.png", "mediaType": "image" }, { "nodeId": 48, "nodeType": "LoadImage", "file": "wooden_doll_back.png", "mediaType": "image" } ] } }, { "name": "3d_triposplat_image_to_gaussian_splat", "title": "TripoSplat：图像到高斯泼溅", "description": "上传一张2D图像。生成具有可控密度和渲染预算的高质量3D高斯溅射表示。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "3D", "图生3D模型" ], "models": [ "TripoSplat" ], "date": "2026-06-02", "openSource": true, "size": 3972844749, "vram": 3972844749, "usage": 0, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 99, "nodeType": "LoadImage", "file": "white-hotel-on-rocky-island.png", "mediaType": "image" } ] } }, { "name": "api_meshy_image_to_model", "title": "Meshy 6：图像转模型", "description": "上传图像并使用 Meshy 6 生成 3D 模型。创建具有生产质量几何结构和干净拓扑的角色、物体或机械部件。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "3D", "图生模型", "API" ], "models": [ "Meshy" ], "date": "2026-01-19", "openSource": false, "usage": 83, "io": { "inputs": [ { "nodeId": 8, "nodeType": "LoadImage", "file": "robot-warrior.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_rodin3d_gen2_5_text_to_3d", "title": "Rodin Gen2.5: Text to Model", "description": "Input a text description or upload reference images. Generate a detailed 3D model with textures, exportable in multiple formats.", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "3D", "API", "文生模型" ], "models": [ "Rodin" ], "logos": [ { "provider": "Rodin" } ], "date": "2026-05-23", "openSource": false, "size": 0, "vram": 0, "usage": 6, "searchRank": 0, "username": "ComfyUI" }, { "name": "api_meshy_multi_image_to_model", "title": "Meshy 6：多图像生成模型", "description": "使用 Meshy 6 从多张图像创建高质量 3D 模型。照片越多，细节捕捉越精细，比例越准确，网格结构越清晰。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "3D", "API" ], "models": [ "Meshy" ], "date": "2026-01-19", "openSource": false, "size": 0, "vram": 0, "usage": 36, "searchRank": 0, "io": { "inputs": [ { "nodeId": 8, "nodeType": "LoadImage", "file": "joker_front.jpg", "mediaType": "image" }, { "nodeId": 12, "nodeType": "LoadImage", "file": "joker_back.jpg", "mediaType": "image" }, { "nodeId": 13, "nodeType": "LoadImage", "file": "joker_side.jpg", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_hunyuan3d_part", "title": "HY 3D：3D零件分解", "description": "上传3D模型以自动将其分割为组成部分，生成完全分解的3D资产以供重用和编辑。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "3D", "3D模型", "API" ], "models": [ "Hunyuan3D" ], "logos": [ { "provider": "Hunyuan" } ], "date": "2026-02-13", "openSource": false, "size": 0, "vram": 0, "usage": 6, "searchRank": 0, "username": "ComfyUI" }, { "name": "api_hunyuan3d_text_to_model", "title": "HunYuan3D：文本生成模型", "description": "输入文本提示或上传参考图像以生成详细的3D模型资源。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "3D", "文生模型" ], "models": [ "Hunyuan3D" ], "date": "2026-01-26", "openSource": false, "usage": 11, "username": "ComfyUI" }, { "name": "api_hunyuan3d_smart_topology", "title": "混元3D：智能拓扑", "description": "上传高多边形3D模型文件。生成指定简化级别的低多边形、拓扑优化的3D模型。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "3D模型" ], "models": [ "Hunyuan3D" ], "logos": [ { "provider": "Hunyuan" } ], "date": "2026-03-07", "openSource": false, "size": 0, "vram": 0, "usage": 11, "searchRank": 0, "username": "ComfyUI" }, { "name": "api_tripo3_1_text_to_model", "title": "Tripo H3.1：文本生成模型", "description": "生成具有高密度几何体和PBR材质的、可用于生产的3D英雄资产。接受文本提示或参考图像作为输入，并输出适用于特写渲染和游戏资产的详细3D网格。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "3D模型", "API" ], "models": [ "Tripo" ], "logos": [ { "provider": "Tripo" } ], "date": "2026-05-09", "openSource": false, "size": 0, "vram": 0, "usage": 18, "searchRank": 0, "username": "ComfyUI" }, { "name": "api_tripo3_1_multiview_to_model", "title": "Tripo H3.1：多视图到模型", "description": "上传物体的多视图图像。生成具有密集几何结构和PBR材质的高细节3D模型。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "3D模型" ], "models": [ "Tripo" ], "logos": [ { "provider": "Tripo" } ], "date": "2026-05-09", "openSource": false, "size": 0, "vram": 0, "usage": 44, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 44, "nodeType": "LoadImage", "file": "castle_front_view.png", "mediaType": "image" }, { "nodeId": 45, "nodeType": "LoadImage", "file": "castle_side_view.png", "mediaType": "image" }, { "nodeId": 46, "nodeType": "LoadImage", "file": "castle_back_view.png", "mediaType": "image" } ] } }, { "name": "api_tripo3_0_image_to_model", "title": "Tripo3.0：图生模型", "description": "将图像或素描转换为3D模型，Tripo 3.0具备锐利的几何结构和可用于生产的PBR材质。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生模型", "3D", "API" ], "models": [ "Tripo" ], "date": "2025-12-23", "openSource": false, "size": 0, "vram": 0, "usage": 21, "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadImage", "file": "3d_character_front_view.png", "mediaType": "image" }, { "nodeId": 3, "nodeType": "LoadImage", "file": "3d_character_back_view.png", "mediaType": "image" }, { "nodeId": 4, "nodeType": "LoadImage", "file": "3d_character_left_view.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "templates_mjm_image_to_3d", "title": "从单张图像生成3D模型", "description": "输入单张图像，自动生成三视图。使用这些多视图与Meshy创建带纹理的3D模型。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "3D模型", "图生3D模型" ], "models": [ "Meshy", "Nano Banana 2" ], "logos": [ { "provider": [ "Google", "Meshy" ] } ], "date": "2026-03-19", "openSource": false, "size": 0, "vram": 0, "usage": 39, "searchRank": 0, "io": { "inputs": [ { "nodeId": 16, "nodeType": "LoadImage", "file": "input_storyboard_video.jpg", "mediaType": "image" } ] }, "thumbnail": [ "thumbnail/templates_mjm_image_to_3d-1.png", "thumbnail/templates_mjm_image_to_3d-2.png" ], "username": "Julien | MJM" }, { "name": "api_meshy_text_to_model", "title": "Meshy 6：文生模型", "description": "使用 Meshy 6 从文本提示生成高质量 3D 模型。创建具有精细几何结构的角色、带有锐利边缘的机械物体，或游戏就绪的低多边形资源。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "3D", "文生模型", "API" ], "models": [ "Meshy" ], "date": "2026-01-19", "openSource": false, "size": 0, "vram": 0, "usage": 3, "searchRank": 0, "username": "ComfyUI" }, { "name": "api_tripo3_0_text_to_model", "title": "Tripo3.0：文生模型", "description": "从文本生成高精度3D模型，具备超清几何和真实PBR材质。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生模型", "3D", "API" ], "models": [ "Tripo" ], "date": "2025-12-23", "openSource": false, "size": 0, "vram": 0, "usage": 5, "username": "ComfyUI" }, { "name": "api_tripo_text_to_model", "title": "Tripo：文本到模型", "description": "使用Tripo的文本驱动建模从描述中制作3D对象。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文生模型", "3D", "API" ], "models": [ "Tripo" ], "date": "2025-03-01", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 6, "username": "ComfyUI" }, { "name": "api_tripo_image_to_model", "title": "Tripo：图像到模型", "description": "使用Tripo引擎从2D图像生成专业3D资产。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "3D", "图生模型", "API" ], "models": [ "Tripo" ], "date": "2025-03-01", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 20, "io": { "inputs": [ { "nodeId": 36, "nodeType": "LoadImage", "file": "api_tripo_image_to_model_input_image.jpg", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_tripo_multiview_to_model", "title": "Tripo：多视角到模型", "description": "使用Tripo的高级扫描仪从多个角度构建3D模型。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生3D模型", "3D", "API" ], "models": [ "Tripo" ], "date": "2025-03-01", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 9, "io": { "inputs": [ { "nodeId": 7, "nodeType": "LoadImage", "file": "api_tripo_multiview_to_model_front_image.jpg", "mediaType": "image" }, { "nodeId": 8, "nodeType": "LoadImage", "file": "api_tripo_multiview_to_model_back_image.jpg", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_rodin_gen2", "title": "Rodin: Gen-2 图生3D模型", "description": "使用 Rodin Gen2 从照片生成详细的4X网格质量3D模型", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生3D模型", "3D", "API" ], "models": [ "Rodin" ], "date": "2025-09-27", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 15, "io": { "inputs": [ { "nodeId": 82, "nodeType": "LoadImage", "file": "api_rodin_gen2_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_rodin_image_to_model", "title": "Rodin：图像到模型", "description": "使用Rodin AI从单张照片生成详细的3D模型。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生3D模型", "3D", "API" ], "models": [ "Rodin" ], "date": "2025-03-01", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 2, "io": { "inputs": [ { "nodeId": 49, "nodeType": "LoadImage", "file": "api_rodin_image_to_model_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_rodin_multiview_to_model", "title": "Rodin：多视角到模型", "description": "使用Rodin的多角度重建雕刻全面的3D模型。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图生3D模型", "3D", "API" ], "models": [ "Rodin" ], "date": "2025-03-01", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 2, "io": { "inputs": [ { "nodeId": 40, "nodeType": "LoadImage", "file": "api_rodin_multiview_to_model_left.jpg", "mediaType": "image" }, { "nodeId": 41, "nodeType": "LoadImage", "file": "api_rodin_multiview_to_model_front.jpg", "mediaType": "image" }, { "nodeId": 56, "nodeType": "LoadImage", "file": "api_rodin_multiview_to_model_back.jpg", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "3d_hunyuan3d-v2.1", "title": "混元3D 2.1", "mediaType": "image", "mediaSubtype": "webp", "description": "使用 Hunyuan3D 2.0 从单张图像生成 3D 模型。", "tags": [ "图生3D模型", "3D" ], "models": [ "Hunyuan3D", "Tencent" ], "date": "2025-03-01", "tutorialUrl": "", "size": 4928474972, "vram": 4928474972, "usage": 104, "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadImage", "file": "3d_hunyuan3d-v2.1_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "3d_hunyuan3d_image_to_model", "title": "混元3D 2.0", "mediaType": "image", "mediaSubtype": "webp", "description": "使用Hunyuan3D 2.0从单张图像生成3D模型。", "tags": [ "图生3D模型", "3D" ], "models": [ "Hunyuan3D", "Tencent" ], "date": "2025-03-01", "tutorialUrl": "", "size": 4928474972, "vram": 4928474972, "usage": 15, "io": { "inputs": [ { "nodeId": 56, "nodeType": "LoadImage", "file": "3d_hunyuan3d_image_to_model_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "3d_hunyuan3d_multiview_to_model", "title": "混元3D 2.0 多视图", "mediaType": "image", "mediaSubtype": "webp", "description": "使用Hunyuan3D 2.0 MV从多个视角生成3D模型。", "tags": [ "3D", "图生3D模型" ], "models": [ "Hunyuan3D", "Tencent" ], "date": "2025-03-01", "tutorialUrl": "", "thumbnailVariant": "hoverDissolve", "size": 4928474972, "vram": 4928474972, "usage": 16, "io": { "inputs": [ { "nodeId": 56, "nodeType": "LoadImage", "file": "3d_hunyuan3d_multiview_to_model_front_image.png", "mediaType": "image" }, { "nodeId": 78, "nodeType": "LoadImage", "file": "3d_hunyuan3d_multiview_to_model_left_image.png", "mediaType": "image" }, { "nodeId": 80, "nodeType": "LoadImage", "file": "3d_hunyuan3d_multiview_to_model_back_image.png", "mediaType": "image" }, { "nodeId": 87, "nodeType": "LoadImage", "file": "3d_hunyuan3d_multiview_to_model_right_image.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "3d_hunyuan3d_multiview_to_model_turbo", "title": "混元3D 2.0 多视图 Turbo", "mediaType": "image", "mediaSubtype": "webp", "description": "使用Hunyuan3D 2.0 MV Turbo从多个视角生成3D模型。", "tags": [ "图生3D模型", "3D" ], "models": [ "Hunyuan3D", "Tencent" ], "date": "2025-03-01", "tutorialUrl": "", "thumbnailVariant": "hoverDissolve", "size": 4928474972, "vram": 4928474972, "usage": 24, "io": { "inputs": [ { "nodeId": 56, "nodeType": "LoadImage", "file": "3d_hunyuan3d_multiview_to_model_turbo_front_view.png", "mediaType": "image" }, { "nodeId": 82, "nodeType": "LoadImage", "file": "3d_hunyuan3d_multiview_to_model_turbo_back_view.png", "mediaType": "image" }, { "nodeId": 85, "nodeType": "LoadImage", "file": "3d_hunyuan3d_multiview_to_model_turbo_left_view.png", "mediaType": "image" }, { "nodeId": 87, "nodeType": "LoadImage", "file": "3d_hunyuan3d_multiview_to_model_turbo_right_view.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_hunyuan3d_retopo_uv", "title": "混元3D：模型重拓扑与UV展开", "description": "输入密集网格（来自AI生成或摄影测量扫描），创建具有良好拓扑结构和展开UV的轻量模型。非常适合3D游戏资产优化。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "3D模型" ], "models": [ "Hunyuan3D" ], "logos": [ { "provider": "Hunyuan" } ], "date": "2026-03-07", "openSource": false, "size": 0, "vram": 0, "usage": 17, "searchRank": 0, "username": "shane" }, { "name": "api_hunyuan3d_model2uv", "title": "混元3D：UV展开", "description": "上传3D模型以执行UV展开。生成具有优化UV布局的已处理模型，用于纹理贴图。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "3D模型" ], "models": [ "Hunyuan3D" ], "logos": [ { "provider": "Hunyuan" } ], "date": "2026-03-07", "openSource": false, "size": 0, "vram": 0, "usage": 6, "searchRank": 0, "username": "ComfyUI", "thumbnail": [ "thumbnail/api_hunyuan3d_model2uv-1.png", "thumbnail/api_hunyuan3d_model2uv-2.png" ] }, { "name": "3d_moge_panorama_to_mesh", "title": "Moge：全景到网格", "description": "上传一张等距柱状投影的360°全景图像，生成带有顶点颜色的纹理化GLB网格。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "3D", "图生3D模型" ], "models": [ "MoGe" ], "date": "2026-05-19", "openSource": true, "size": 644245094, "vram": 644245094, "usage": 47, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 28, "nodeType": "LoadImage", "file": "lego_street_panorama.png", "mediaType": "image" } ] } }, { "name": "3d_moge_perspective_to_mesh", "title": "MoGe：透视几何估计", "description": "上传图像以估算其透视几何。从输入生成3D深度图和表面法线。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "3D", "图生3D模型" ], "models": [ "MoGe" ], "date": "2026-05-15", "openSource": true, "size": 644245094, "vram": 644245094, "usage": 120, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 9, "nodeType": "LoadImage", "file": "modern_living_room.png", "mediaType": "image" } ] } } ] }, { "moduleName": "default", "type": "image", "category": "生成类型", "icon": "icon-[lucide--message-square-text]", "title": "LLM", "templates": [ { "name": "api_openrouter_llm", "title": "OpenRouter LLM ", "description": "从OpenRouter精选列表中选择一个模型（Claude、GPT、Gemini等）。生成文本响应，并可选择上传媒体文件以用于支持视觉的模型。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文本生成", "API" ], "models": [ "OpenRouter" ], "logos": [ { "provider": "OpenRouter" } ], "date": "2026-05-21", "openSource": false, "size": 0, "vram": 0, "usage": 75, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadImage", "file": "contact_sheet_app_input_1.png", "mediaType": "image" } ] } }, { "name": "api_anthropic_claude", "title": "Anthropic Claude", "description": "输入您的查询或图像进行分析。生成由Anthropic Claude驱动的自然语言响应，支持文本对话和图像理解。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "文本生成" ], "models": [ "Claude" ], "logos": [ { "provider": "Anthropic" } ], "date": "2026-05-01", "openSource": false, "size": 0, "vram": 0, "usage": 229, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadImage", "file": "ice_cold_pup.png", "mediaType": "image" } ] } }, { "name": "llm_gemma4_text_gen", "tutorialUrl": "https://docs.comfy.org/tutorials/llm/gemma4/gemma4", "title": "Gemm4：文本生成", "description": "输入您的文本提示，并可选择图像、音频或视频。生成具有可配置推理、编码和多语言支持的文本输出。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文本生成", "文本生成" ], "models": [ "Gemma 4" ], "date": "2026-05-01", "openSource": true, "size": 9019431322, "vram": 9019431322, "usage": 147, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadImage", "file": "the_lily_veil.png", "mediaType": "image" }, { "nodeId": 5, "nodeType": "LoadAudio", "file": "voice_demo.mp3", "mediaType": "audio" }, { "nodeId": 6, "nodeType": "LoadVideo", "file": "video_wan_vace_inpainting_input_video.mp4", "mediaType": "video" } ] } }, { "name": "llm_qwen3_5_text_gen", "tutorialUrl": "https://docs.comfy.org/tutorials/llm/qwen/qwen3_5", "title": "Qwen3.5：文本生成", "description": "使用Qwen3.5模型分析输入图像并生成描述性文本提示。此工作流程执行图像描述和反向提示工程。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文本生成" ], "models": [ "Qwen3.5" ], "date": "2026-03-27", "openSource": true, "size": 9341553869, "vram": 9341553869, "usage": 105, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadImage", "file": "man_with_red_hat.png", "mediaType": "image" } ] }, "thumbnail": [ "thumbnail/llm_qwen3_5_text_gen.png" ] }, { "name": "llm_qwen3_text_gen", "tutorialUrl": "https://docs.comfy.org/tutorials/llm/qwen/qwen3", "title": "Qwen3.0：文本生成", "description": "输入文本提示，使用Qwen3-4B-Thinking模型生成详细、经过推理的回复。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "文本生成" ], "models": [ "Qwen 3.0" ], "date": "2026-02-23", "openSource": true, "size": 8053063680, "vram": 8053063680, "usage": 65, "searchRank": 0, "username": "ComfyUI", "thumbnail": [ "thumbnail/llm_qwen3_text_gen.png" ] }, { "name": "api_openai_chat", "title": "OpenAI：聊天", "description": "与OpenAI的高级语言模型互动进行智能对话。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "文本生成" ], "models": [ "OpenAI" ], "date": "2025-03-01", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 84, "io": { "inputs": [ { "nodeId": 5, "nodeType": "LoadImage", "file": "example.png", "mediaType": "image" } ] }, "username": "ComfyUI", "searchRank": 0, "logos": [ { "provider": "OpenAI" } ] }, { "name": "api_google_gemini", "title": "Google Gemini", "description": "体验Google的多模态AI和Gemini的推理能力。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "文本生成" ], "models": [ "Google", "Google Gemini" ], "date": "2025-03-01", "tutorialUrl": "", "openSource": false, "size": 0, "vram": 0, "usage": 41, "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadImage", "file": "example.png", "mediaType": "image" } ] }, "username": "ComfyUI", "searchRank": 0, "logos": [ { "provider": "Google" } ] }, { "name": "api_bytedance_seed", "title": "字节跳动：Seed 2.0", "description": "输入图像或视频进行分析。接收描述视觉内容、推理和结构化信息提取的详细文本输出。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API" ], "models": [ "Seed 2.0" ], "logos": [ { "provider": "ByteDance" } ], "date": "2026-05-19", "openSource": false, "size": 0, "vram": 0, "usage": 72, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 3, "nodeType": "LoadImage", "file": "angle_3.png", "mediaType": "image" } ] } } ] }, { "moduleName": "default", "type": "image", "isEssential": true, "category": "生成类型", "icon": "icon-[lucide--settings]", "title": "工具", "templates": [ { "name": "api_beeble_switchx_video_edit", "title": "Beeble SwitchX：视频编辑", "description": "上传源视频和参考图像，然后提供描述所需环境和光照的提示。生成编辑后的视频，其中遮罩区域遵循您的提示和参考，而未遮罩区域保留原始像素和运动，并附带用于合成的Alpha遮罩。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "视频编辑" ], "models": [ "SwitchX" ], "logos": [ { "provider": "Beeble" } ], "date": "2026-05-29", "openSource": false, "size": 0, "vram": 0, "usage": 0, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 17, "nodeType": "LoadImage", "file": "blue_poppies_field.png", "mediaType": "image" }, { "nodeId": 21, "nodeType": "LoadVideo", "file": "woman_in_wheat_field.mp4", "mediaType": "video" } ] } }, { "name": "api_beeble_switchx_image_edit", "title": "Beeble SwitchX：图像编辑", "description": "上传您的角色图像和参考风格/光照图像。生成将参考光照和风格应用于原始图像的编辑输出。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "API", "重打光", "图片编辑" ], "models": [ "SwitchX" ], "logos": [ { "provider": "Beeble" } ], "date": "2026-05-29", "openSource": false, "size": 0, "vram": 0, "usage": 0, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 10, "nodeType": "LoadImage", "file": "soft_light_headphones_man.png", "mediaType": "image" }, { "nodeId": 17, "nodeType": "LoadImage", "file": "simple_modern_living_space.png", "mediaType": "image" } ] } }, { "name": "api_flux_erase_image", "title": "Flux 擦除图像", "description": "上传一张图片，并在要移除的对象上绘制遮罩。工作流将擦除遮罩区域并重建背景。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "图像修复" ], "models": [ "Flux" ], "logos": [ { "provider": "Black Forest Labs" } ], "date": "2026-06-04", "openSource": false, "size": 0, "vram": 0, "usage": 0, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 10, "nodeType": "LoadImage", "file": "city_street_young_man.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 3, "nodeType": "SaveImage", "file": "api_flux_erase_image.png", "mediaType": "image" } ] }, "thumbnail": [ "input/city_street_young_man.png", "output/api_flux_erase_image.png" ] }, { "name": "api_bria_remove_video_background", "title": "Bria：移除视频背景", "description": "上传视频以移除其背景，并替换为您选择的颜色。生成一个主体分离且带有新纯色背景的视频。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "移除背景" ], "models": [ "BRIA" ], "logos": [ { "provider": "BRIA" } ], "date": "2026-06-04", "openSource": false, "size": 0, "vram": 0, "usage": 0, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 4, "nodeType": "LoadVideo", "file": "spear_warrior.mp4", "mediaType": "video" } ] } }, { "name": "api_bria_remove_video_background_transparent", "title": "Bria：移除视频背景（透明）", "description": "上传视频并生成具有透明背景的帧，输出图像和遮罩序列，同时支持Alpha通道的WebM视频。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "移除背景" ], "models": [ "BRIA" ], "logos": [ { "provider": "BRIA" } ], "date": "2026-06-04", "openSource": false, "size": 0, "vram": 0, "usage": 0, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 5, "nodeType": "LoadVideo", "file": "ad_video_demo.mp4", "mediaType": "video" } ] } }, { "name": "api_flux_vto", "title": "FLUX 虚拟试穿", "description": "上传一张人物照片和一张服装图片。生成虚拟试穿效果，将服装应用到人物身上，同时保留面部和姿势。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "虚拟试穿" ], "models": [ "Flux" ], "logos": [ { "provider": "Black Forest Labs" } ], "date": "2026-06-04", "openSource": false, "size": 0, "vram": 0, "usage": 0, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 1, "nodeType": "LoadImage", "file": "portrait_floral_headshot.png", "mediaType": "image" }, { "nodeId": 5, "nodeType": "LoadImage", "file": "green_chiffon_short_dress.png", "mediaType": "image" } ] } }, { "name": "api_bria_video_green_screen", "title": "Bria：视频绿幕", "description": "上传视频以生成带有透明或纯色背景的绿幕版本。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "移除背景", "API" ], "models": [ "BRIA" ], "logos": [ { "provider": "BRIA" } ], "date": "2026-06-04", "openSource": false, "size": 0, "vram": 0, "usage": 0, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 5, "nodeType": "LoadVideo", "file": "investigator.mp4", "mediaType": "video" } ] } }, { "name": "utility_void_video_inpainting", "tutorialUrl": "https://docs.comfy.org/tutorials/utility/void-video-inpainting", "title": "VOID: Video Inpaint", "description": "上传视频并遮罩要移除的对象。生成一个干净的视频，其中该对象及其物理交互已被删除。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "视频编辑", "视频修复" ], "models": [ "VOID" ], "date": "2026-05-14", "openSource": true, "size": 34252364186, "vram": 34252364186, "usage": 112, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 4, "nodeType": "LoadVideo", "file": "snowboarder.mp4", "mediaType": "video" } ], "outputs": [ { "nodeId": 38, "nodeType": "SaveVideo", "file": "utility_void_video_inpainting.mp4", "mediaType": "video" }, { "nodeId": 48, "nodeType": "SaveVideo", "file": "utility_void_video_inpainting.mp4", "mediaType": "video" } ] } }, { "name": "api_topaz_astra2", "title": "Topaz Astra 2：创意扩散视频放大", "description": "使用Astra 2的创意扩散技术提升GenAI视频素材，添加动态细节和纹理。调整创意性、清晰度和基于提示的指导，精确风格化AI生成的内容。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "视频超分辨率" ], "models": [ "Topaz" ], "logos": [ { "provider": "Topaz" } ], "date": "2026-05-03", "openSource": false, "size": 0, "vram": 0, "usage": 85, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 1, "nodeType": "LoadVideo", "file": "cyber_fight.mp4", "mediaType": "video" } ] } }, { "name": "utility_sdpose_ood_video_to_pose_map", "title": "SDPose-OOD：视频到姿态映射", "description": "上传视频以提取姿态关键点并生成姿态图。该工作流程支持多人检测，并使用增强的SDPose模型进行精确的全身特征提取。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "姿态地图" ], "models": [ "SDPose-OOD" ], "date": "2026-03-02", "openSource": true, "size": 1932735283, "vram": 1932735283, "usage": 4, "searchRank": 0, "io": { "inputs": [ { "nodeId": 679, "nodeType": "LoadVideo", "file": "man_in_the_rain.mp4", "mediaType": "video" } ], "outputs": [ { "nodeId": 680, "nodeType": "SaveVideo", "file": "utility_sdpose_ood_video_to_pose_map.mp4", "mediaType": "video" } ] }, "username": "ComfyUI", "thumbnail": [ "input/man_in_the_rain.mp4", "output/utility_sdpose_ood_video_to_pose_map.mp4" ] }, { "name": "utility_birefnet_remove_background", "tutorialUrl": "https://docs.comfy.org/tutorials/utility/remove-background-birefnet", "title": "BiRefNet：背景移除", "description": "上传任意背景的图像。生成一个去除背景的版本以及精确的分割掩码。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "移除背景" ], "models": [ "BiRefNet" ], "date": "2026-05-10", "openSource": true, "size": 429496730, "vram": 429496730, "usage": 108, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 17, "nodeType": "LoadImage", "file": "the_lily_veil.png", "mediaType": "image" } ] } }, { "name": "utility_moge_depth_estimation", "tutorialUrl": "https://docs.comfy.org/tutorials/utility/moge", "title": "MoGe：深度估计", "description": "上传单个RGB图像，调整推理分辨率和批量大小。生成彩色深度预览和原始深度图。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "深度图" ], "models": [ "MoGe" ], "date": "2026-05-20", "openSource": true, "size": 644245094, "vram": 644245094, "usage": 23, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 9, "nodeType": "LoadImage", "file": "alien_world.png", "mediaType": "image" } ] } }, { "name": "utility_video_frame_interpolation", "tutorialUrl": "https://docs.comfy.org/tutorials/utility/frame-interpolation", "title": "视频帧插值", "description": "上传视频并选择RIFE或FILM帧插值模型，以生成更流畅、更高帧率的视频输出。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "视频补帧" ], "models": [ "FILM", "RIFE" ], "date": "2026-04-27", "openSource": true, "size": 107374182, "vram": 107374182, "usage": 14, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 4, "nodeType": "LoadVideo", "file": "white_armored_warrior.mp4", "mediaType": "video" } ], "outputs": [ { "nodeId": 7, "nodeType": "SaveVideo", "file": "utility_video_frame_interpolation.mp4", "mediaType": "video" } ] }, "thumbnail": [ "input/white_armored_warrior.mp4", "output/utility_video_frame_interpolation.mp4" ] }, { "name": "utility_sdpose_multi_person_video", "title": "SDPose：视频多人检测", "description": "上传视频以检测人体姿态。支持在同一场景中检测单人和多人。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "姿态地图" ], "models": [ "SDPose-OOD" ], "date": "2026-03-31", "openSource": true, "size": 2040109466, "vram": 2040109466, "usage": 3, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 694, "nodeType": "LoadVideo", "file": "man_playing_violin.mp4", "mediaType": "video" } ] } }, { "name": "utility_sdpose_multi_person", "title": "SDPose：图像多人检测", "description": "上传图像以检测人体姿态。支持检测单个人物以及同一场景中的多个人物。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "姿态地图" ], "models": [ "SDPose-OOD" ], "date": "2026-03-31", "openSource": true, "size": 2040109466, "vram": 2040109466, "usage": 1, "searchRank": 0, "username": "ComfyUI", "thumbnail": [ "thumbnail/utility_sdpose_multi_person.mp4" ], "io": { "inputs": [ { "nodeId": 679, "nodeType": "LoadImage", "file": "group_photo.png", "mediaType": "image" } ] } }, { "name": "utility_sdpose_ood_image_to_pose", "title": "SDPose-OOD：图像到姿态映射", "description": "上传图像以提取姿态关键点，并使用SDPose-OOD模型生成相应的姿态图。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "姿态地图" ], "models": [ "SDPose-OOD" ], "date": "2026-03-02", "openSource": true, "size": 1932735283, "vram": 1932735283, "usage": 48, "searchRank": 0, "io": { "inputs": [ { "nodeId": 667, "nodeType": "LoadImage", "file": "dancer.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 669, "nodeType": "SaveImage", "file": "utility_sdpose_ood_image_to_pose.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "input/dancer.png", "output/utility_sdpose_ood_image_to_pose.png" ] }, { "name": "utility_pid_latent_upscale_dit", "title": "PiD：潜在放大解码", "description": "从扩散模型输入潜在图像并选择PiD检查点。通过单次4倍或8倍放大生成高分辨率解码图像。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "图像放大" ], "models": [ "PiD" ], "date": "2026-05-19", "openSource": true, "includeOnDistributions": [ "local" ], "size": 25984552141, "vram": 25984552141, "usage": 0, "searchRank": 0, "username": "ComfyUI" }, { "name": "utility_image_segment_sam3", "tutorialUrl": "https://docs.comfy.org/tutorials/utility/video-segment-sam3", "title": "SAM3：图像分割", "description": "使用SAM3模型从照片或图像中分割主要主体或内容，隔离特定对象或区域。输入图像并接收分割后的掩码输出。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "元素分割" ], "models": [ "SAM3" ], "date": "2026-04-26", "openSource": true, "size": 1717986918, "vram": 1717986918, "usage": 87, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 79, "nodeType": "LoadImage", "file": "neon_guitarist.png", "mediaType": "image" } ] }, "thumbnail": [ "thumbnail/utility_image_segment_sam3.mp4" ] }, { "name": "api_magnific_image_upscale_creative", "title": "Magnific：创意图像放大", "description": "上传图片并设置放大级别。生成支持高达16倍放大的图像高清版本。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "图像放大" ], "models": [ "Magnific" ], "logos": [ { "provider": "Magnific" } ], "date": "2026-02-13", "openSource": false, "size": 0, "vram": 0, "usage": 138, "searchRank": 0, "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadImage", "file": "1950_new_york.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 3, "nodeType": "SaveImage", "file": "ComfyUI_00915_.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "input/1950_new_york.png", "output/ComfyUI_00915_.png" ] }, { "name": "api_topaz_starlight_precise25", "title": "视频放大：Topaz Starlight Precise 2.5", "description": "上传视频，使用Topaz Starlight Precise 2.5模型提升其分辨率。此工作流程相比先前版本能生成更清晰、伪影更少的4K输出。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "视频超分辨率", "API" ], "models": [ "Topaz" ], "logos": [ { "provider": "Topaz" } ], "date": "2026-03-26", "openSource": false, "size": 0, "vram": 0, "usage": 81, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 3, "nodeType": "LoadVideo", "file": "running_horse.mp4", "mediaType": "video" } ], "outputs": [ { "nodeId": 4, "nodeType": "SaveVideo", "file": "running_horse.mp4", "mediaType": "video" } ] }, "thumbnail": [ "thumbnail/api_topaz_starlight_precise25.mp4" ] }, { "name": "utility_z_image_turbo_2k_upscaler.app", "tutorialUrl": "https://docs.comfy.org/tutorials/utility/image-upscale", "title": "图像放大：Z-Image-Turbo 2K", "description": "上传图片，使用Z-Image-Turbo模型将其提升至2K分辨率。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "图像放大" ], "models": [ "Z-Image-Turbo" ], "date": "2026-02-19", "openSource": true, "size": 20723217203, "vram": 20723217203, "usage": 494, "searchRank": 0, "io": { "inputs": [ { "nodeId": 77, "nodeType": "LoadImage", "file": "fighter_jet.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 9, "nodeType": "SaveImage", "file": "utility_z_image_turbo_2k_upscaler.png", "mediaType": "image" } ] }, "username": "hellorob", "thumbnail": [ "input/fighter_jet.png", "output/utility_z_image_turbo_2k_upscaler.png" ] }, { "name": "utility_bria_remove_image_background", "title": "Bira：移除图片背景", "description": "上传图片，使用Bria API自动移除背景。工作流程异步处理请求，并返回带有透明背景的编辑后图片。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "移除背景", "图像" ], "models": [ "BRIA" ], "logos": [ { "provider": "BRIA" } ], "date": "2026-02-16", "openSource": false, "size": 0, "vram": 0, "usage": 46, "searchRank": 0, "io": { "inputs": [ { "nodeId": 26, "nodeType": "LoadImage", "file": "denim_girl.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 27, "nodeType": "SaveImage", "file": "utility_bria_remove_image_background.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "input/denim_girl.png", "output/utility_bria_remove_image_background.png" ] }, { "name": "utility-bria_remove_video_background", "title": "Bria：移除视频背景", "description": "上传视频以移除背景。生成具有透明或纯色背景的新视频。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "移除背景" ], "models": [ "BRIA" ], "logos": [ { "provider": "BRIA" } ], "date": "2026-02-16", "openSource": false, "size": 0, "vram": 0, "usage": 12, "searchRank": 0, "io": { "inputs": [ { "nodeId": 123, "nodeType": "LoadVideo", "file": "prince_and_phoenix.mp4", "mediaType": "video" } ], "outputs": [ { "nodeId": 124, "nodeType": "SaveVideo", "file": "utility-bria_remove_video_background.mp4", "mediaType": "video" } ] }, "username": "ComfyUI", "thumbnail": [ "input/prince_and_phoenix.mp4", "output/utility-bria_remove_video_background.mp4" ] }, { "name": "api_magnific_image_upscale_precise", "title": "Magnific：精准图像放大", "description": "上传图像以使用Magnific模型精确放大，生成具有增强细节的高分辨率版本。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "图像放大" ], "models": [ "Magnific" ], "logos": [ { "provider": "Magnific" } ], "date": "2026-02-13", "openSource": false, "size": 0, "vram": 0, "usage": 101, "searchRank": 0, "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadImage", "file": "young_woman_portrait.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 3, "nodeType": "SaveImage", "file": "api_magnific_image_upscale_precise.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "input/young_woman_portrait.png", "output/api_magnific_image_upscale_precise.png" ] }, { "name": "utility-topaz_landscape_upscaler", "title": "放大：使用 Topaz Reimagine 进行风景图像放大", "description": "上传一张风景图片进行放大和细节增强。使用Topaz Reimagine生成针对自然风光优化的高分辨率输出。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "图像放大" ], "models": [ "Topaz" ], "date": "2026-02-02", "openSource": false, "size": 0, "vram": 0, "usage": 36, "searchRank": 0, "logos": [ { "provider": "Topaz" } ], "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadImage", "file": "moonlit_cliff_house.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_topaz_image_enhance", "title": "Topaz： 图像增强", "description": "使用 Topaz 的 Reimagine 模型进行专业图像增强，同时增强人脸并恢复图像细节。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "图像放大", "API" ], "models": [ "Reimagine", "Topaz" ], "date": "2025-11-25", "openSource": false, "size": 0, "vram": 0, "usage": 66, "searchRank": 0, "logos": [ { "provider": "Topaz" } ], "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadImage", "file": "api_topaz_image_enhance_input_image.webp", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "utility_nanobanana_pro_ai_image_fix", "title": "Upscale：使用 Nano Banana Pro 进行 AI 图像增强", "description": "上传一张AI生成的图像进行放大和细节增强。使用Nano Banana Pro生成具有更高清晰度和一致性的4K输出。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "图像放大" ], "models": [ "Google", "Nano Banana Pro" ], "date": "2026-02-02", "openSource": false, "size": 0, "vram": 0, "usage": 208, "searchRank": 0, "logos": [ { "provider": "Google" } ], "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadImage", "file": "hovering_figure.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 5, "nodeType": "SaveImage", "file": "utility_nanobanana_pro_ai_image_fix.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "input/hovering_figure.png", "output/utility_nanobanana_pro_ai_image_fix.png" ] }, { "name": "api_topaz_video_enhance", "title": "Topaz Astra：视频创意增强", "description": "使用 Topaz AI 增强视频。支持通过 Starlight（Astra）Fast 模型进行分辨率放大，并支持通过 apo-8 模型进行画面插帧。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "视频超分辨率" ], "models": [ "Topaz" ], "date": "2025-11-25", "openSource": false, "size": 0, "vram": 0, "usage": 33, "searchRank": 0, "io": { "inputs": [ { "nodeId": 8, "nodeType": "LoadVideo", "file": "api_topaz_video_enhance_input_video.mp4", "mediaType": "video" } ] }, "username": "ComfyUI" }, { "name": "utility_topaz_illustration_upscale", "title": "放大：使用 Topaz 进行插画放大", "description": "上传插图以使用Topaz进行放大。生成平滑、风格化的图像，色彩准确，但可能会出现一些伪影。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "图像放大" ], "models": [ "Topaz" ], "date": "2026-02-02", "openSource": false, "size": 0, "vram": 0, "usage": 29, "searchRank": 0, "logos": [ { "provider": "Topaz" } ], "io": { "inputs": [ { "nodeId": 5, "nodeType": "LoadImage", "file": "silent_sanctuary.jpg", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_stability_upscale_fast", "title": "Stability AI：快速图像放大", "description": "上传图片，通过Stability AI API快速将其放大4倍，非常适合提升低质量或压缩图片的画质。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "图像放大" ], "models": [ "Stability" ], "logos": [ { "provider": "Stability" } ], "date": "2026-02-26", "openSource": false, "size": 0, "vram": 0, "usage": 13, "searchRank": 0, "io": { "inputs": [ { "nodeId": 9, "nodeType": "LoadImage", "file": "woman_with_red_hat.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 4, "nodeType": "SaveImage", "file": "api_stability_upscale_fast.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "input/woman_with_red_hat.png", "output/api_stability_upscale_fast.png" ] }, { "name": "utility_nanobanana_pro_illustration_upscale", "title": "Upscale：使用Nano Banana Pro进行风格化艺术创意放大", "description": "上传风格化艺术或插图。使用Nano Banana Pro模型生成创意升级的4K图像，增强细节，适合营销或游戏素材。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "图像放大" ], "models": [ "Gemini3 Pro Image Preview", "Google", "Nano Banana Pro" ], "date": "2026-02-02", "openSource": false, "size": 0, "vram": 0, "usage": 35, "searchRank": 0, "logos": [ { "provider": "Google" } ], "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadImage", "file": "downhill_run.jpg", "mediaType": "image" } ], "outputs": [ { "nodeId": 5, "nodeType": "SaveImage", "file": "utility_nanobanana_pro_illustration_upscale.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "input/downhill_run.jpg", "output/utility_nanobanana_pro_illustration_upscale.png" ] }, { "name": "utility_nanobanana_pro_product_upscale", "title": "画质提升：使用Nano Banana Pro增强产品细节", "description": "上传产品图片以增强其细节和清晰度。使用Nano Banana Pro模型生成放大、高质量的输出，具有改进的排版和逼真的连贯性。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "图像放大", "产品" ], "models": [ "Nano Banana Pro" ], "date": "2026-02-02", "openSource": false, "size": 0, "vram": 0, "usage": 26, "searchRank": 0, "logos": [ { "provider": "Google" } ], "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadImage", "file": "orange_beauty_trio.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "utility_hitpaw_video_enhance", "title": "HitPaw：视频增强与升级", "description": "上传视频以增强和放大。生成具有更清晰画质、时间稳定性和可选人像增强的修复输出。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "视频超分辨率", "API" ], "models": [ "HitPaw" ], "logos": [ { "provider": "HitPaw" } ], "date": "2026-02-02", "openSource": false, "size": 0, "vram": 0, "usage": 10, "searchRank": 0, "io": { "inputs": [ { "nodeId": 12, "nodeType": "LoadVideo", "file": "dad_carry_daughter.mp4", "mediaType": "video" } ] }, "username": "ComfyUI" }, { "name": "utility_hitpaw_general_image_enhance", "title": "HitPaw：人像扩散增强器", "description": "上传一张低质量的人像图片并调整放大系数。生成高保真、细节丰富的人像，增强皮肤纹理和面部特征。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "图像放大" ], "models": [ "HitPaw" ], "logos": [ { "provider": "HitPaw" } ], "date": "2026-02-02", "openSource": false, "size": 0, "vram": 0, "usage": 17, "searchRank": 0, "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadImage", "file": "sunlit_girl_portrait.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 3, "nodeType": "SaveImage", "file": "utility_hitpaw_general_image_enhance.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "input/sunlit_girl_portrait.png", "output/utility_hitpaw_general_image_enhance.png" ] }, { "name": "utility_recraft_creative_image_upscale", "title": "Recraft：创意图像增强放大工具", "description": "上传图像以增强和放大。Recraft的AI重建细节并改善纹理，输出具有自然外观的更高质量图像。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "图像放大" ], "models": [ "Recraft" ], "date": "2026-02-02", "openSource": false, "size": 0, "vram": 0, "usage": 13, "searchRank": 0, "logos": [ { "provider": "Recraft" } ], "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadImage", "file": "butterfly_girl.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 3, "nodeType": "SaveImage", "file": "utility_recraft_creative_image_upscale.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "input/butterfly_girl.png", "output/utility_recraft_creative_image_upscale.png" ] }, { "name": "utility_recraft_crisp_image_upscale", "title": "Recraft：清晰图像超分辨率修复", "description": "上传低分辨率或噪点图像进行修复和放大。生成高分辨率输出，增强细节，减少伪影，保持自然外观。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "图像放大" ], "models": [ "Recraft" ], "date": "2026-02-02", "openSource": false, "size": 0, "vram": 0, "usage": 27, "searchRank": 0, "logos": [ { "provider": "Recraft" } ], "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadImage", "file": "man_with_rose.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 3, "nodeType": "SaveImage", "file": "utility_recraft_crisp_image_upscale.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "input/man_with_rose.png", "output/utility_recraft_crisp_image_upscale.png" ] }, { "name": "api_wavespeed_seedvr2_ai_image_fix", "title": "Upscale：使用 WaveSpeed SeedVR2 进行 AI 图像修复", "description": "上传AI生成的图像以恢复细节并放大。工作流程在保留颜色和锐度的同时最小化伪影。输出为高质量、放大后的输入版本。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "图像放大" ], "models": [ "SeedVR2", "WaveSpeed" ], "date": "2026-02-02", "openSource": false, "size": 0, "vram": 0, "usage": 32, "searchRank": 0, "logos": [ { "provider": "WaveSpeed" } ], "io": { "inputs": [ { "nodeId": 9, "nodeType": "LoadImage", "file": "midnight_coastal.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 3, "nodeType": "SaveImage", "file": "api_wavespeed_seedvr2_ai_image_fix.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "input/midnight_coastal.png", "output/api_wavespeed_seedvr2_ai_image_fix.png" ] }, { "name": "api_wavespeed_image_upscale", "title": "WaveSpeed：图像放大", "description": "上传图像以提升其分辨率。生成增强的2K、4K或8K输出，保留细节并减少JPEG、PNG或WEBP格式的伪影。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "图像放大" ], "models": [ "WaveSpeed" ], "date": "2026-01-23", "openSource": false, "size": 0, "vram": 0, "usage": 34, "searchRank": 0, "logos": [ { "provider": "WaveSpeed" } ], "io": { "inputs": [ { "nodeId": 9, "nodeType": "LoadImage", "file": "woman_behind_veil.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 3, "nodeType": "SaveImage", "file": "api_wavespeed_image_upscale.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "input/woman_behind_veil.png", "output/api_wavespeed_image_upscale.png" ] }, { "name": "api_wavespeed_flshvsr_video_upscale", "title": "WaveSpeed：FlashVSR 视频超分辨率", "description": "上传视频以将其升级至720p、1080p、2K或4K。该工作流程可增强细节、减少伪影并确保运动一致，从而获得高质量、自然观感的输出。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "视频超分辨率" ], "models": [ "FlashVSR", "WaveSpeed" ], "date": "2026-01-23", "openSource": false, "usage": 35, "size": 0, "vram": 0, "searchRank": 0, "logos": [ { "provider": "WaveSpeed" } ], "io": { "inputs": [ { "nodeId": 4, "nodeType": "LoadVideo", "file": "lighter.mp4", "mediaType": "video" } ] }, "username": "ComfyUI" }, { "name": "image_qwen_image_layered", "tutorialUrl": "https://docs.comfy.org/tutorials/image/qwen/qwen-image-layered", "title": "Qwen-Image-Layered 图层分解", "mediaType": "image", "mediaSubtype": "webp", "description": "分解图像为可编辑的RGBA", "tags": [ "图层分解" ], "models": [ "Qwen-Image-Layered" ], "date": "2025-12-22", "size": 50465865728, "vram": 50465865728, "usage": 46, "openSource": true, "searchRank": 0, "io": { "inputs": [ { "nodeId": 74, "nodeType": "LoadImage", "file": "coastal_smiling_woman.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "image_qwen_image_layered_control", "tutorialUrl": "https://docs.comfy.org/tutorials/image/qwen/qwen-image-layered", "title": "Qwen-Image-Layered 控制分解", "description": "描述并提取图像元素。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图层分解" ], "models": [ "Qwen-Image-Layered" ], "date": "2026-01-15", "openSource": true, "usage": 17, "io": { "inputs": [ { "nodeId": 74, "nodeType": "LoadImage", "file": "raccoon_rocker.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "api_vidu_video_extension", "title": "Vidu：视频扩展", "description": "上传视频并使用Vidu Q2模型将其延长至7秒，并将分辨率提升至1080p。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "视频", "视频扩展", "API" ], "models": [ "Vidu Q2", "Vidu" ], "date": "2026-02-01", "openSource": false, "logos": [ { "provider": "Vidu" } ], "usage": 6, "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadVideo", "file": "lawyer.mp4", "mediaType": "video" } ] }, "username": "ComfyUI" }, { "name": "api_magnific_image_relight", "title": "Magnific：图像重光", "description": "上传源图像和参考光照图像。根据参考图像的光照条件，生成源图像的重新光照版本。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "重打光" ], "models": [ "Magnific" ], "date": "2026-01-26", "openSource": false, "usage": 12, "io": { "inputs": [ { "nodeId": 11, "nodeType": "LoadImage", "file": "magical-portrait.png", "mediaType": "image" }, { "nodeId": 13, "nodeType": "LoadImage", "file": "vintage-girl.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 14, "nodeType": "SaveImage", "file": "api_magnific_image_relight.png", "mediaType": "image" } ] }, "username": "ComfyUI", "logos": [ { "provider": "Magnific" } ], "size": 0, "vram": 0, "searchRank": 0, "thumbnail": [ "input/magical-portrait.png", "output/api_magnific_image_relight.png" ] }, { "name": "api_magnific_skin_enhancer", "title": "Magnific：皮肤增强", "description": "通过上传您的图像并选择预设来调整光照、色彩分级和真实感，从而提升皮肤和整体图像质量。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "图像增强", "肖像" ], "models": [ "Magnific" ], "date": "2026-01-26", "openSource": false, "usage": 111, "size": 0, "vram": 0, "searchRank": 0, "io": { "inputs": [ { "nodeId": 4, "nodeType": "LoadImage", "file": "close_up_girl_skin.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 5, "nodeType": "SaveImage", "file": "api_magnific_skin_enhancer.png", "mediaType": "image" } ] }, "username": "ComfyUI", "logos": [ { "provider": "Magnific" } ], "thumbnail": [ "input/close_up_girl_skin.png", "output/api_magnific_skin_enhancer.png" ] }, { "name": "api_bfl_flux1_expand_image", "title": "Flux.1：图像扩展", "description": "上传图像以使用 Flux.1 扩展其画布。生成带有扩展内容的向外绘制图像。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "API", "外绘" ], "models": [ "Flux", "Flux.1" ], "date": "2026-02-09", "openSource": false, "size": 0, "vram": 0, "usage": 15, "searchRank": 0, "io": { "inputs": [ { "nodeId": 17, "nodeType": "LoadImage", "file": "man_on_desert.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 5, "nodeType": "SaveImage", "file": "api_bfl_flux1_expand_image.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "thumbnail/api_bfl_flux1_expand_image-1.png", "thumbnail/api_bfl_flux1_expand_image-2.png" ] }, { "name": "utility-gan_upscaler", "tutorialUrl": "https://docs.comfy.org/tutorials/utility/image-upscale", "title": "快速GAN视频增强器", "description": "上传视频以使用快速GAN模型进行超分辨率处理。输出分辨率由您选择的模型决定（例如2倍或4倍）。为获得最佳效果，请先进行超分辨率处理，然后调整至目标分辨率。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "视频超分辨率", "API" ], "models": [ "Real-ESRGAN" ], "date": "2026-01-26", "openSource": true, "usage": 46, "io": { "inputs": [ { "nodeId": 9, "nodeType": "LoadVideo", "file": "gan_input.mp4", "mediaType": "video" } ] }, "username": "ComfyUI" }, { "name": "image_lotus_depth_v1_1", "tutorialUrl": "https://docs.comfy.org/tutorials/utility/moge", "title": "Lotus深度", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "description": "在ComfyUI中运行Lotus深度，实现零样本、高效的单目深度估计，并保持高细节保留。", "tags": [ "图像", "文生图" ], "models": [ "SD1.5", "Stability" ], "date": "2025-05-21", "size": 2040109466, "vram": 2040109466, "usage": 31, "io": { "inputs": [ { "nodeId": 12, "nodeType": "LoadImage", "file": "image_lotus_depth_v1_1_input_image.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 9, "nodeType": "SaveImage", "file": "image_lotus_depth_v1_1.png", "mediaType": "image" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0, "thumbnail": [ "input/image_lotus_depth_v1_1_input_image.png", "output/image_lotus_depth_v1_1.png" ] }, { "name": "utility_interpolation_image_upscale", "tutorialUrl": "https://docs.comfy.org/tutorials/utility/image-upscale", "title": "图像放大：传统插值", "description": "上传图像并选择插值方法。使用传统算法放大图像，以实现更快的处理速度，同时保持细节变化最小。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "图像放大" ], "models": [ "None" ], "date": "2026-02-01", "openSource": true, "size": 0, "vram": 0, "usage": 22, "searchRank": 0, "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadImage", "file": "pray.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 5, "nodeType": "SaveImage", "file": "utility_interpolation_image_upscale.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "input/pray.png", "output/utility_interpolation_image_upscale.png" ] }, { "name": "utility_video_segment_sam3", "tutorialUrl": "https://docs.comfy.org/tutorials/utility/video-segment-sam3", "title": "SAM3：视频分割", "description": "使用SAM3模型从视频中分割出主要主体或内容，隔离特定对象或区域。输入视频并接收分割后的掩码输出。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "元素分割" ], "models": [ "SAM3" ], "date": "2026-04-27", "openSource": true, "size": 1717986918, "vram": 1717986918, "usage": 34, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 115, "nodeType": "LoadVideo", "file": "drinking_unicorn.mp4", "mediaType": "video" } ] }, "thumbnail": [ "output/utility_video_segment_sam3.mp4" ] } ] }, { "moduleName": "default", "type": "image", "isEssential": true, "icon": "icon-[lucide--graduation-cap]", "title": "快速开始", "templates": [ { "name": "gsl_starter_1_1", "title": "1.1 入门 – 文生图", "description": "学习ComfyUI中图像生成的基础知识 — 加载模型、编写提示词并生成您的第一张图像。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图像", "文生图" ], "models": [ "SD1.5" ], "date": "2026-02-12", "openSource": true, "includeOnDistributions": [ "local" ], "size": 2147483648, "vram": 2147483648, "usage": 12, "searchRank": 0, "username": "ComfyUI" }, { "name": "gsl_starter_1_2", "title": "1.2 入门 – 图生视频", "description": "了解如何使用 Wan 2.2 加载图像、生成视频以及查找节点", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "视频", "首尾帧视频" ], "models": [ "Wan2.2" ], "date": "2026-02-10", "openSource": true, "includeOnDistributions": [ "local" ], "size": 36829344563, "vram": 36829344563, "usage": 1, "searchRank": 0, "io": { "inputs": [ { "nodeId": 122, "nodeType": "LoadImage", "file": "gs2_image1.png", "mediaType": "image" }, { "nodeId": 124, "nodeType": "LoadImage", "file": "gs2_image2.png", "mediaType": "image" }, { "nodeId": 126, "nodeType": "LoadImage", "file": "gs2_image3.png", "mediaType": "image" }, { "nodeId": 130, "nodeType": "LoadImage", "file": "gs2_image4.png", "mediaType": "image" } ] }, "username": "ComfyUI" }, { "name": "gsl_starter_1_3", "title": "1.3 入门 – 产品摄影", "description": "了解如何使用Qwen Image Edit重新打光产品、进入子图、取消节点旁路。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图像", "图片编辑" ], "models": [ "Flux.2 Klein" ], "date": "2026-02-08", "openSource": true, "includeOnDistributions": [ "local" ], "size": 12455405158, "vram": 12455405158, "usage": 7, "searchRank": 0, "io": { "inputs": [ { "nodeId": 3, "nodeType": "LoadImage", "file": "gsc3_landscape.png", "mediaType": "image" }, { "nodeId": 7, "nodeType": "LoadImage", "file": "gsc3_car.jpeg", "mediaType": "image" } ], "outputs": [ { "nodeId": 2, "nodeType": "SaveImage", "file": "gsl_starter_1_2.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnail": [ "output/gsl_starter_1_2.png" ] }, { "name": "gsl_creator_2", "title": "2.2 Creator - Image Editing", "description": "加载图像，在要编辑的区域上绘制遮罩，并输入提示。生成一个新图像，其中只有遮罩区域根据您的提示重新绘制。", "mediaType": "image", "mediaSubtype": "webp", "thumbnailVariant": "compareSlider", "tags": [ "图像修复" ], "models": [ "Z-Image-Turbo" ], "date": "2026-02-05", "openSource": true, "includeOnDistributions": [ "local" ], "size": 27380416512, "vram": 27380416512, "usage": 0, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 137, "nodeType": "LoadImage", "file": "old_school_style_girl.png", "mediaType": "image" } ] } }, { "name": "01_get_started_text_to_image", "title": "文生图", "mediaType": "image", "mediaSubtype": "webp", "description": "使用 z-image-turbo 模型根据文本提示生成图像", "tutorialUrl": "https://docs.comfy.org/tutorials/image/z-image/z-image-turbo", "tags": [ "图像", "文生图" ], "models": [ "Z-Image-Turbo" ], "date": "2025-10-17", "size": 20830591386, "vram": 20830591386, "usage": 114, "io": { "outputs": [ { "nodeId": 60, "nodeType": "SaveImage", "file": "01_get_started_text_to_image.png", "mediaType": "image" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0, "thumbnail": [ "output/01_get_started_text_to_image.png" ] }, { "name": "02_qwen_Image_edit_subgraphed", "title": "图像编辑", "mediaType": "image", "mediaSubtype": "webp", "description": "使用 Qwen-Image-Edit 编辑你的图片", "tutorialUrl": "https://docs.comfy.org/tutorials/image/qwen/qwen-image-edit", "tags": [ "ControlNet", "图片编辑" ], "models": [ "Qwen-Image-Edit" ], "date": "2025-10-17", "size": 31782757990, "vram": 31782757990, "usage": 1168, "io": { "inputs": [ { "nodeId": 78, "nodeType": "LoadImage", "file": "02_qwen_Image_edit_subgraphed_input_image.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 60, "nodeType": "SaveImage", "file": "02_qwen_Image_edit_subgraphed.png", "mediaType": "image" } ] }, "username": "ComfyUI", "thumbnailVariant": "compareSlider", "openSource": true, "searchRank": 0, "thumbnail": [ "input/02_qwen_Image_edit_subgraphed_input_image.png", "output/02_qwen_Image_edit_subgraphed.png" ] }, { "name": "03_video_wan2_2_14B_i2v_subgraphed", "title": "图生视频", "description": "使用 Wan2.2 14B 从输入图像生成视频", "mediaType": "image", "mediaSubtype": "webp", "tutorialUrl": "https://docs.comfy.org/tutorials/video/wan/wan2_2", "tags": [ "图生视频", "视频" ], "models": [ "Wan", "Wan2.2" ], "date": "2025-10-17", "size": 38010460570, "vram": 38010460570, "usage": 815, "io": { "inputs": [ { "nodeId": 97, "nodeType": "LoadImage", "file": "03_video_wan2_2_14B_i2v_subgraphed_input_image.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 108, "nodeType": "SaveVideo", "file": "03_video_wan2_2_14B_i2v_subgraphed.mp4", "mediaType": "video" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0, "thumbnail": [ "output/03_video_wan2_2_14B_i2v_subgraphed.mp4" ] }, { "name": "04_hunyuan_3d_2.1_subgraphed", "title": "图生3D", "mediaType": "image", "mediaSubtype": "webp", "description": "使用 Hunyuan3D 2.1 从单张图片生成 3D 模型。", "tags": [ "图生3D模型", "3D" ], "models": [ "Hunyuan3D" ], "date": "2025-10-17", "tutorialUrl": "https://docs.comfy.org/tutorials/3d/hunyuan3D-2", "size": 4939212390, "vram": 4939212390, "usage": 93, "io": { "inputs": [ { "nodeId": 2, "nodeType": "LoadImage", "file": "04_hunyuan_3d_2.1_subgraphed_input_image.png", "mediaType": "image" } ] }, "username": "ComfyUI", "openSource": true, "searchRank": 0 }, { "name": "05_audio_ace_step_1_t2a_song_subgraphed", "title": "文生音频", "mediaType": "image", "mediaSubtype": "webp", "description": "使用ACE-Step v1根据文本提示生成音频", "tags": [ "音频", "文本转音频" ], "models": [ "ACE-Step" ], "date": "2025-10-17", "tutorialUrl": "https://docs.comfy.org/tutorials/audio/ace-step/ace-step-v1", "size": 7730941133, "vram": 7730941133, "usage": 1, "username": "ComfyUI", "openSource": true, "searchRank": 0 }, { "name": "default", "title": "图像生成", "mediaType": "image", "mediaSubtype": "webp", "description": "通过文本提示生成图像。", "tutorialUrl": "https://docs.comfy.org/tutorials/basic/text-to-image", "tags": [ "文生图", "图像" ], "models": [ "SD1.5", "Stability" ], "date": "2025-03-01", "size": 2147483648, "vram": 3113851290, "usage": 115, "username": "ComfyUI", "openSource": true, "searchRank": 0 } ] }, { "moduleName": "default", "type": "image", "isEssential": true, "category": "生成类型", "icon": "icon-[lucide--blocks]", "title": "节点基础", "templates": [ { "name": "basic_image_color_adjustment", "title": "色彩调整蓝图展示", "description": "内置蓝图颜色分级子图的并排演示。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "节点基础" ], "models": [ "None" ], "date": "2026-04-24", "openSource": true, "size": 0, "vram": 0, "usage": 21, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 180, "nodeType": "LoadImage", "file": "example.png", "mediaType": "image" } ] } }, { "name": "basic_datatype_conversion", "title": "内置数据类型转换示例", "description": "演示如何在工作流中转换各种内置数据类型。输入任意数据值并输出其转换后的形式。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "节点基础" ], "models": [ "None" ], "date": "2026-03-24", "openSource": true, "size": 0, "vram": 0, "usage": 5, "searchRank": 0, "username": "ComfyUI" }, { "name": "basic_mask_operations_and_compositing", "title": "遮罩操作、合并与合成", "description": "核心图像和遮罩节点的交互式演练 — 创建和优化遮罩，通过不同操作组合重叠遮罩，并使用羽化遮罩合成图像。无需模型。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "节点基础" ], "models": [ "None" ], "date": "2026-03-24", "openSource": true, "size": 0, "vram": 0, "usage": 21, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 374, "nodeType": "LoadImage", "file": "example.png", "mediaType": "image" } ] }, "thumbnail": [ "thumbnail/basic_mask_operations_and_compositing.png" ] }, { "name": "basic_switch_node", "title": "切换节点", "description": "Switch Nodes 工作原理指南，简单文本切换与 LoRA 切换", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "节点基础" ], "models": [ "Z-Image-Turbo" ], "date": "2026-03-30", "openSource": true, "size": 20723217203, "vram": 20723217203, "usage": 19, "searchRank": 0, "username": "ComfyUI" }, { "name": "utility_image_stitch", "title": "图像拼接 2x2 网格", "description": "将四张输入图像组合成2x2网格。自动调整不匹配图像的尺寸并输出一张2048x2048的图像。", "mediaType": "image", "mediaSubtype": "webp", "tags": [ "图像" ], "models": [ "None" ], "date": "2026-03-19", "openSource": true, "size": 0, "vram": 0, "usage": 5, "searchRank": 0, "username": "ComfyUI", "io": { "inputs": [ { "nodeId": 12, "nodeType": "LoadImage", "file": "angle_1.png", "mediaType": "image" }, { "nodeId": 13, "nodeType": "LoadImage", "file": "angle_2.png", "mediaType": "image" }, { "nodeId": 14, "nodeType": "LoadImage", "file": "angle_3.png", "mediaType": "image" }, { "nodeId": 15, "nodeType": "LoadImage", "file": "angle_4.png", "mediaType": "image" } ], "outputs": [ { "nodeId": 19, "nodeType": "SaveImage", "file": "utility_image_stitch.png", "mediaType": "image" } ] }, "thumbnail": [ "thumbnail/utility_image_stitch.mp4" ] } ] } ]