3j SSKrSSKJrJr SSKrSSKJrJrJrJ r SSK J r SSK J r Jr SSKJrJr SSKJr SS KJr SS KJr SS KJrJrJrJrJrJr SS KJ r S SK!J"r" S SK#J$r$ SSK%J&r& \"5(a SSK'J(s J)r* Sr+OSr+\RX"\-5r.Sr/SS\0S-S\1\Rd-S-S\3\0S-S\3\4S-4Sjjr5"SS\"\\ \&5r6g)N)AnyCallable)CLIPTextModelWithProjection CLIPTokenizerT5EncoderModelT5TokenizerFast)VaeImageProcessor)FromSingleFileMixinSD3LoraLoaderMixin)PAGCFGJointAttnProcessor2_0PAGJointAttnProcessor2_0) AutoencoderKL)SD3Transformer2DModel)FlowMatchEulerDiscreteScheduler)USE_PEFT_BACKENDis_torch_xla_availableloggingreplace_example_docstringscale_lora_layersunscale_lora_layers) randn_tensor)DiffusionPipeline)StableDiffusion3PipelineOutput)PAGMixinTFa^ Examples: ```py >>> import torch >>> from diffusers import AutoPipelineForText2Image >>> pipe = AutoPipelineForText2Image.from_pretrained( ... "stabilityai/stable-diffusion-3-medium-diffusers", ... torch_dtype=torch.float16, ... enable_pag=True, ... pag_applied_layers=["blocks.13"], ... ) >>> pipe.to("cuda") >>> prompt = "A cat holding a sign that says hello world" >>> image = pipe(prompt, guidance_scale=5.0, pag_scale=0.7).images[0] >>> image.save("sd3_pag.png") ``` num_inference_stepsdevice timestepssigmasc UbUb [S5eUbS[[R"UR5R R 55;nU(d[SURS35eUR"S X2S.UD6 URn[U5nX14$UbS[[R"UR5R R 55;nU(d[SURS35eUR"S XBS.UD6 URn[U5nX14$UR"U4S U0UD6 URnX14$) a Calls the scheduler's `set_timesteps` method and retrieves timesteps from the scheduler after the call. Handles custom timesteps. Any kwargs will be supplied to `scheduler.set_timesteps`. Args: scheduler (`SchedulerMixin`): The scheduler to get timesteps from. num_inference_steps (`int`): The number of diffusion steps used when generating samples with a pre-trained model. If used, `timesteps` must be `None`. device (`str` or `torch.device`, *optional*): The device to which the timesteps should be moved to. If `None`, the timesteps are not moved. timesteps (`list[int]`, *optional*): Custom timesteps used to override the timestep spacing strategy of the scheduler. If `timesteps` is passed, `num_inference_steps` and `sigmas` must be `None`. sigmas (`list[float]`, *optional*): Custom sigmas used to override the timestep spacing strategy of the scheduler. If `sigmas` is passed, `num_inference_steps` and `timesteps` must be `None`. Returns: `tuple[torch.Tensor, int]`: A tuple where the first element is the timestep schedule from the scheduler and the second element is the number of inference steps. zYOnly one of `timesteps` or `sigmas` can be passed. Please choose one to set custom valuesr zThe current scheduler class zx's `set_timesteps` does not support custom timestep schedules. Please check whether you are using the correct scheduler.)r rr!zv's `set_timesteps` does not support custom sigmas schedules. Please check whether you are using the correct scheduler.)r!rr) ValueErrorsetinspect signature set_timesteps parameterskeys __class__r len) schedulerrrr r!kwargsaccepts_timesteps accept_sigmass c/home/wildlama/miniconda3/lib/python3.13/site-packages/diffusers/pipelines/pag/pipeline_pag_sd_3.pyretrieve_timestepsr2Ms}>!3tuu'3w/@/@AXAX/Y/d/d/i/i/k+ll .y/B/B.CDab  M)MfM'' !)n  ))   C(9(9):Q:Q(R(](](b(b(d$ee .y/B/B.CD_`  GvGG'' !)n  ))  3MFMfM''  ))c9z^\rSrSrSrSr/r/SQrSFS\S\ S\ S\ S \ S \ S \ S \ S \S\\\-4U4SjjjrSGS\\\-S\S\S\R(S-S\R*S-4 SjjrSHS\\\-S\S\R(S-S\S-S\4 SjjrSIS\\\-S\\\-S\\\-S\R(S-S\S\S \\\-S-S!\\\-S-S"\\\-S-S#\R2S-S$\R2S-S%\R2S-S&\R2S-S\S-S\S'\S-4 S(jjrSJS)jrSKS*jr\S+5r\S,5r \S-5r!\S.5r"\S/5r#\S05r$\RJ"5\&"\'5SSSSSS1SS2SSSSSSSSSSS3SSSSS4/SS5S64S\\\-S\\\-S-S\\\-S-S7\S-S8\S-S9\S:\\S-S;\S \\\-S-S!\\\-S-S"\\\-S-S\S-S<\RP\\RP-S-S4\R2S-S#\R2S-S$\R2S-S%\R2S-S&\R2S-S=\S-S>\S?\)\\*4S-S\S-S@\+\\/S4S-SA\\S\SB\SC\46SDjj55r,SEr-U=r.$)LStableDiffusion3PAGPipelinea\ [PAG pipeline](https://huggingface.co/docs/diffusers/main/en/using-diffusers/pag) for text-to-image generation using Stable Diffusion 3. Args: transformer ([`SD3Transformer2DModel`]): Conditional Transformer (MMDiT) architecture to denoise the encoded image latents. scheduler ([`FlowMatchEulerDiscreteScheduler`]): A scheduler to be used in combination with `transformer` to denoise the encoded image latents. vae ([`AutoencoderKL`]): Variational Auto-Encoder (VAE) Model to encode and decode images to and from latent representations. text_encoder ([`CLIPTextModelWithProjection`]): [CLIP](https://huggingface.co/docs/transformers/model_doc/clip#transformers.CLIPTextModelWithProjection), specifically the [clip-vit-large-patch14](https://huggingface.co/openai/clip-vit-large-patch14) variant, with an additional added projection layer that is initialized with a diagonal matrix with the `hidden_size` as its dimension. text_encoder_2 ([`CLIPTextModelWithProjection`]): [CLIP](https://huggingface.co/docs/transformers/model_doc/clip#transformers.CLIPTextModelWithProjection), specifically the [laion/CLIP-ViT-bigG-14-laion2B-39B-b160k](https://huggingface.co/laion/CLIP-ViT-bigG-14-laion2B-39B-b160k) variant. text_encoder_3 ([`T5EncoderModel`]): Frozen text-encoder. Stable Diffusion 3 uses [T5](https://huggingface.co/docs/transformers/model_doc/t5#transformers.T5EncoderModel), specifically the [t5-v1_1-xxl](https://huggingface.co/google/t5-v1_1-xxl) variant. tokenizer (`CLIPTokenizer`): Tokenizer of class [CLIPTokenizer](https://huggingface.co/docs/transformers/v4.21.0/en/model_doc/clip#transformers.CLIPTokenizer). tokenizer_2 (`CLIPTokenizer`): Second Tokenizer of class [CLIPTokenizer](https://huggingface.co/docs/transformers/v4.21.0/en/model_doc/clip#transformers.CLIPTokenizer). tokenizer_3 (`T5TokenizerFast`): Tokenizer of class [T5Tokenizer](https://huggingface.co/docs/transformers/model_doc/t5#transformers.T5Tokenizer). z>text_encoder->text_encoder_2->text_encoder_3->transformer->vae)latents prompt_embedsnegative_prompt_embedsnegative_pooled_prompt_embeds transformerr-vae text_encoder tokenizertext_encoder_2 tokenizer_2text_encoder_3 tokenizer_3pag_applied_layersc >[T U]5 URUUUUUUU UUS9 [USS5(a/S[ UR R R5S- -OSUl[URS9Ul [US5(a#URbURROSUl[US 5(a-URb URR R OS Ul[US 5(a-URb URR R$OSUlUR'U [)5[+54S 9 g) N) r<r=r?rAr>r@rBr;r-r<rr)vae_scale_factorr>Mr;)pag_attn_processors)super__init__register_modulesgetattrr,r<configblock_out_channelsrFr image_processorhasattrr>model_max_lengthtokenizer_max_lengthr; sample_sizedefault_sample_size patch_sizeset_pag_applied_layersr r) selfr;r-r<r=r>r?r@rArBrCr+s r1rK$StableDiffusion3PAGPipeline.__init__sX  %))###  W^^bdikoVpVpc$((//*L*L&MPQ&Q Rvw0$BWBWX/6t[/I/IdnnNhDNN + +np ! t]++0@0@0L    # # / /  3:$ 2N2NSWScScSoD   # # . .uv  ## 5P5RTlTn4o $ r3Nrpromptnum_images_per_promptmax_sequence_lengthrdtypec U=(d URnU=(d URRn[U[5(aU/OUn[ U5nUR c9[R"Xb-UURRR4UUS9$URUSUSSSS9nURnURUSSS9Rn U RSURS:ag[R"X5(dLURR!U SS2UR"S - S245n [$R'S US U 35 UR UR)U55S n UR RnU R)XTS 9n U Rupn U R+S US 5n U R-Xb-U S5n U $)Nrr^ max_lengthTpt)paddingra truncationadd_special_tokensreturn_tensorslongestrcrfrzXThe following part of your input was truncated because `max_sequence_length` is set to tokens: rr^r)_execution_devicer=r^ isinstancestrr,rAtorchzerosr;rNjoint_attention_dimrB input_idsshapeequal batch_decoderSloggerwarningtorepeatview)rXr[r\r]rr^ batch_size text_inputstext_input_idsuntruncated_ids removed_textr8_seq_lens r1_get_t5_prompt_embeds1StableDiffusion3PAGPipeline._get_t5_prompt_embedss14110**00'44&&[    &;;6'$$++??  &&  *# ' %..**69UY*Zdd   $(<(r@r=r?rmrnr,rSrrrsrortrurvrwrx hidden_statesr^ryrz)rXr[r\rrrclip_tokenizersclip_text_encodersr>r=r{r|r}r~rr8pooled_prompt_embedsrrs r1_get_clip_prompt_embeds3StableDiffusion3PAGPipeline._get_clip_prompt_embedss1411>>4+;+;<"//1D1DE#5 ); '44&&[   00  %..#FIdS]]   $(<(UYZ ,Q/  )77;M)77)a-8HIM%((t/@/@/F/Fv(V %++ 7A%,,Q0EqI %**:+MwXZ[ 3::1>ST3889[]_`222r3Tprompt_2prompt_3do_classifier_free_guidancenegative_promptnegative_prompt_2negative_prompt_3r8r9rr: lora_scalec VU=(d URnUbx[U[5(acUUlURb![ (a[ URU5 URb![ (a[ URU5 [U[5(aU/OUnUb [U5nOU RSnU GcU=(d Un[U[5(aU/OUnU=(d Un[U[5(aU/OUnURUUUUSS9unnURUUUUSS9unn[R"UU/SS9nURUUUUS9n[RR R#USURSURS- 45n[R"UU/SS9n [R"UU/SS9n U(GaU GcU=(d S nU=(d UnU =(d Un [U[5(aUU/-OUn[U[5(aUU/-OUn[U [5(aUU /-OU n Ub;[%U5[%U5La$['S [%U5S [%U5S 35eU[U5:wa![)S US[U5SUSUS3 5eURUUUSSS9unnURUUUSSS9unn[R"UU/SS9nURU UUUS9n[RR R#USURSURS- 45n[R"UU/SS9n [R"UU/SS9n URb6[U[5(a![ (a[+URU5 URb6[U[5(a![ (a[+URU5 XX4$)a Args: prompt (`str` or `list[str]`, *optional*): prompt to be encoded prompt_2 (`str` or `list[str]`, *optional*): The prompt or prompts to be sent to the `tokenizer_2` and `text_encoder_2`. If not defined, `prompt` is used in all text-encoders prompt_3 (`str` or `list[str]`, *optional*): The prompt or prompts to be sent to the `tokenizer_3` and `text_encoder_3`. If not defined, `prompt` is used in all text-encoders device: (`torch.device`): torch device num_images_per_prompt (`int`): number of images that should be generated per prompt do_classifier_free_guidance (`bool`): whether to use classifier free guidance or not negative_prompt (`str` or `list[str]`, *optional*): The prompt or prompts not to guide the image generation. If not defined, one has to pass `negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is less than `1`). negative_prompt_2 (`str` or `list[str]`, *optional*): The prompt or prompts not to guide the image generation to be sent to `tokenizer_2` and `text_encoder_2`. If not defined, `negative_prompt` is used in all the text-encoders. negative_prompt_3 (`str` or `list[str]`, *optional*): The prompt or prompts not to guide the image generation to be sent to `tokenizer_3` and `text_encoder_3`. If not defined, `negative_prompt` is used in all the text-encoders. prompt_embeds (`torch.FloatTensor`, *optional*): Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not provided, text embeddings will be generated from `prompt` input argument. negative_prompt_embeds (`torch.FloatTensor`, *optional*): Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input argument. pooled_prompt_embeds (`torch.FloatTensor`, *optional*): Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not provided, pooled text embeddings will be generated from `prompt` input argument. negative_pooled_prompt_embeds (`torch.FloatTensor`, *optional*): Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt` input argument. clip_skip (`int`, *optional*): Number of layers to be skipped from CLIP while computing the prompt embeddings. A value of 1 means that the output of the pre-final layer will be used for computing the prompt embeddings. lora_scale (`float`, *optional*): A lora scale that will be applied to all LoRA layers of the text encoder if LoRA layers are loaded. Nr)r[rr\rrrridim)r[r\r]rrz?`negative_prompt` should be the same type to `prompt`, but got z != .z`negative_prompt`: z has batch size z, but `prompt`: zT. Please make sure that passed `negative_prompt` matches the batch size of `prompt`.)rr\rr)rlrmr _lora_scaler=rrr?rnr,rsrrocatrnn functionalpadtype TypeErrorr$r)rXr[rrrr\rrrrr8r9rr:rr]rr{ prompt_embedpooled_prompt_embedprompt_2_embedpooled_prompt_2_embedclip_prompt_embedst5_prompt_embednegative_prompt_embednegative_pooled_prompt_embednegative_prompt_2_embednegative_pooled_prompt_2_embednegative_clip_prompt_embedst5_negative_prompt_embeds r1 encode_prompt)StableDiffusion3PAGPipeline.encode_promptOsD1411  !j7I&J&J)D   ,1A1A!$"3"3Z@"".3C3C!$"5"5zB'44&&  VJ&,,Q/J  )6H%/#%>%>zHH)6H%/#%>%>zHH040L0L&;#!" 1M1 -L-594P4P&;#!" 5Q5 1N1"'L.+Ir!R "88&;$7 9O"'!4!4!8!8"Q(=(=b(ADVD\D\]_D`(`$a" "II'9?&KQSTM#(99.ACX-Y_a#b &+A+I-3O 1 D_  1 D_ AK?\_@`@`jO+<?PRU4V4V /00\m 5??PRU4V4V /00\m !d6l$:O&OUVZ[jVkUlmV ~Q(s?33 )/)::J3K_J`ax/ |<33 CGB^B^&;!" C_C ? !#?GKFbFb!&;!" GcG C #%C+0))5JLc4djl*m ''+'A'A(&;$7 (B( $+0((*=*=*A*A+,22269T9Z9Z[]9^^_+ ' &+YY0KMe/fln%o ",1II-/MNTV- )    ($ 2338H8H#D$5$5zB    *$ 2338H8H#D$7$7D6Jiir3c^UTRTR--S:wd UTRTR--S:waj[STRTR-SUSUSXDTRTR--- SXUTRTR--- S3 5eU bW[U4SjU 55(d=[S TRS U Vs/sHoTR;dMUPM sn35eUbU b[S US U S 35eUbU b[SUS U S 35eUbU b[SUS U S 35eUcU c [S5eUbA[ U[ 5(d,[ U[5(d[S[U535eUbA[ U[ 5(d,[ U[5(d[S[U535eUbA[ U[ 5(d,[ U[5(d[S[U535eUbU b[SUSU S 35eUbU b[SUSU S 35eUbU b[SUSU S 35eU bCU b@U RU R:wa&[SU RSU RS35eU bU c [S5eU bU c [S5eUbUS:a[SU35eggs snf)Nrz-`height` and `width` have to be divisible by z but are z and z.You can use height z and width rc3@># UHoTR;v M g7fN)_callback_tensor_inputs).0krXs r1 ;StableDiffusion3PAGPipeline.check_inputs..+sF 7Y!-- -7Ysz2`callback_on_step_end_tensor_inputs` has to be in z , but found zCannot forward both `prompt`: z and `prompt_embeds`: z2. Please make sure to only forward one of the two.z Cannot forward both `prompt_2`: z Cannot forward both `prompt_3`: zeProvide either `prompt` or `prompt_embeds`. Cannot leave both `prompt` and `prompt_embeds` undefined.z2`prompt` has to be of type `str` or `list` but is z4`prompt_2` has to be of type `str` or `list` but is z4`prompt_3` has to be of type `str` or `list` but is z'Cannot forward both `negative_prompt`: z and `negative_prompt_embeds`: z)Cannot forward both `negative_prompt_2`: z)Cannot forward both `negative_prompt_3`: zu`prompt_embeds` and `negative_prompt_embeds` must have the same shape when passed directly, but got: `prompt_embeds` z != `negative_prompt_embeds` zIf `prompt_embeds` are provided, `pooled_prompt_embeds` also have to be passed. Make sure to generate `pooled_prompt_embeds` from the same text encoder that was used to generate `prompt_embeds`.zIf `negative_prompt_embeds` are provided, `negative_pooled_prompt_embeds` also have to be passed. Make sure to generate `negative_pooled_prompt_embeds` from the same text encoder that was used to generate `negative_prompt_embeds`.iz8`max_sequence_length` cannot be greater than 512 but is ) rFrVr$allrrmrnlistrrs)rXr[rrheightwidthrrrr8r9rr:"callback_on_step_end_tensor_inputsr]rs` r1 check_inputs(StableDiffusion3PAGPipeline.check_inputss{$ d++doo= >! C--?@AE?@U@UX\XgXg@g?hhqrxqyy~@EFF&&,9N9NQUQ`Q`9`/a&a%bbmns@D@U@UX\XgXg@gwhohniijk  . 9#F 7YF C C DTEaEaDbbn|^pH|^vwkokGkGbGpq|^pHoIJ   -";08N}o^00  !m&?28*Z+GA#i.AQR&<'gi  u&Vr3cUR$r_guidance_scalerXs r1guidance_scale*StableDiffusion3PAGPipeline.guidance_scales###r3cUR$r) _clip_skiprs r1r%StableDiffusion3PAGPipeline.clip_skip r3c URS:$)Nrrrs r1r7StableDiffusion3PAGPipeline.do_classifier_free_guidances##a''r3cUR$r)_joint_attention_kwargsrs r1joint_attention_kwargs2StableDiffusion3PAGPipeline.joint_attention_kwargss+++r3cUR$r)_num_timestepsrs r1 num_timesteps)StableDiffusion3PAGPipeline.num_timestepss"""r3cUR$r) _interruptrs r1 interrupt%StableDiffusion3PAGPipeline.interruptrr3g@pilr7g@grrrr!rr output_type return_dictrcallback_on_step_endr pag_scalepag_adaptive_scalec^ U=(d URUR-nU=(d URUR-nURUUUUUU U U UUUUUUS9 XlUUlUUlSUlUUlUUlUb[U[5(aSnO3Ub![U[5(a [U5nOURSnURnURbURR!SS5OSnUR#UUUU U U UR$UUUUUUR&U UUS9unnnnUR((a;UR+UUUR$5nUR+UUUR$5nO?UR$(a.[,R."UU/SS9n[,R."UU/SS9n[0(aS nOUn[3UR4UUUS 9un n[7[U 5X`R4R8-- S5n![U 5UlUR<R>R@n"URCUU -U"UUURDUU U5nUR((a:UR<RFn#URIURJUR$S 9 URMUS 9n$[OU 5GH\un%n&URP(aM[,R."U/URSURS--5n'U&RSU'RS5n(UR=U'U(UUURSS 9Sn)UR((a)URUU)UR$URVU&5n)O:UR$(a)U)RYS5un*n+U*URVU+U*- --n)URDn,UR4R[U)U&USS9SnURDU,:wa>[,R\R^Ra5(aURcU,5nUbn0n-UHn.[e5U.U-U.'M U"UU%U&U-5n/U/RgSU5nU/RgSU5nU/RgSU5nU/RgSU5nU%[U 5S- :Xd)U%S-U!:a0U%S-UR4R8-S:XaU$Ri5 [0(dGMG[jRl"5 GM_ SSS5 US:XaUn0OzXRnR>Rp- URnR>Rr-nURnRuUSS9Sn0URvRyU0US9n0UR{5 UR((aUR<R}W#5 U(dU04$[U0S9$!,(df  N=f)a Function invoked when calling the pipeline for generation. Args: prompt (`str` or `list[str]`, *optional*): The prompt or prompts to guide the image generation. If not defined, one has to pass `prompt_embeds`. instead. prompt_2 (`str` or `list[str]`, *optional*): The prompt or prompts to be sent to `tokenizer_2` and `text_encoder_2`. If not defined, `prompt` is will be used instead prompt_3 (`str` or `list[str]`, *optional*): The prompt or prompts to be sent to `tokenizer_3` and `text_encoder_3`. If not defined, `prompt` is will be used instead height (`int`, *optional*, defaults to self.unet.config.sample_size * self.vae_scale_factor): The height in pixels of the generated image. This is set to 1024 by default for the best results. width (`int`, *optional*, defaults to self.unet.config.sample_size * self.vae_scale_factor): The width in pixels of the generated image. This is set to 1024 by default for the best results. num_inference_steps (`int`, *optional*, defaults to 50): The number of denoising steps. More denoising steps usually lead to a higher quality image at the expense of slower inference. sigmas (`list[float]`, *optional*): Custom sigmas to use for the denoising process with schedulers which support a `sigmas` argument in their `set_timesteps` method. If not defined, the default behavior when `num_inference_steps` is passed will be used. guidance_scale (`float`, *optional*, defaults to 7.0): Guidance scale as defined in [Classifier-Free Diffusion Guidance](https://huggingface.co/papers/2207.12598). `guidance_scale` is defined as `w` of equation 2. of [Imagen Paper](https://huggingface.co/papers/2205.11487). Guidance scale is enabled by setting `guidance_scale > 1`. Higher guidance scale encourages to generate images that are closely linked to the text `prompt`, usually at the expense of lower image quality. negative_prompt (`str` or `list[str]`, *optional*): The prompt or prompts not to guide the image generation. If not defined, one has to pass `negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is less than `1`). negative_prompt_2 (`str` or `list[str]`, *optional*): The prompt or prompts not to guide the image generation to be sent to `tokenizer_2` and `text_encoder_2`. If not defined, `negative_prompt` is used instead negative_prompt_3 (`str` or `list[str]`, *optional*): The prompt or prompts not to guide the image generation to be sent to `tokenizer_3` and `text_encoder_3`. If not defined, `negative_prompt` is used instead num_images_per_prompt (`int`, *optional*, defaults to 1): The number of images to generate per prompt. generator (`torch.Generator` or `list[torch.Generator]`, *optional*): One or a list of [torch generator(s)](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make generation deterministic. latents (`torch.FloatTensor`, *optional*): Pre-generated noisy latents, sampled from a Gaussian distribution, to be used as inputs for image generation. Can be used to tweak the same generation with different prompts. If not provided, a latents tensor will be generated by sampling using the supplied random `generator`. prompt_embeds (`torch.FloatTensor`, *optional*): Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not provided, text embeddings will be generated from `prompt` input argument. negative_prompt_embeds (`torch.FloatTensor`, *optional*): Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input argument. pooled_prompt_embeds (`torch.FloatTensor`, *optional*): Pre-generated pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not provided, pooled text embeddings will be generated from `prompt` input argument. negative_pooled_prompt_embeds (`torch.FloatTensor`, *optional*): Pre-generated negative pooled text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not provided, pooled negative_prompt_embeds will be generated from `negative_prompt` input argument. output_type (`str`, *optional*, defaults to `"pil"`): The output format of the generate image. Choose between [PIL](https://pillow.readthedocs.io/en/stable/): `PIL.Image.Image` or `np.array`. return_dict (`bool`, *optional*, defaults to `True`): Whether or not to return a [`~pipelines.stable_diffusion_xl.StableDiffusionXLPipelineOutput`] instead of a plain tuple. joint_attention_kwargs (`dict`, *optional*): A kwargs dictionary that if specified is passed along to the `AttentionProcessor` as defined under `self.processor` in [diffusers.models.attention_processor](https://github.com/huggingface/diffusers/blob/main/src/diffusers/models/attention_processor.py). callback_on_step_end (`Callable`, *optional*): A function that calls at the end of each denoising steps during the inference. The function is called with the following arguments: `callback_on_step_end(self: DiffusionPipeline, step: int, timestep: int, callback_kwargs: Dict)`. `callback_kwargs` will include a list of all tensors as specified by `callback_on_step_end_tensor_inputs`. callback_on_step_end_tensor_inputs (`list`, *optional*): The list of tensor inputs for the `callback_on_step_end` function. The tensors specified in the list will be passed as `callback_kwargs` argument. You will only be able to include variables listed in the `._callback_tensor_inputs` attribute of your pipeline class. max_sequence_length (`int` defaults to 256): Maximum sequence length to use with the `prompt`. pag_scale (`float`, *optional*, defaults to 3.0): The scale factor for the perturbed attention guidance. If it is set to 0.0, the perturbed attention guidance will not be used. pag_adaptive_scale (`float`, *optional*, defaults to 0.0): The adaptive scale factor for the perturbed attention guidance. If it is set to 0.0, `pag_scale` is used. Examples: Returns: [`~pipelines.stable_diffusion_3.StableDiffusion3PipelineOutput`] or `tuple`: [`~pipelines.stable_diffusion_3.StableDiffusion3PipelineOutput`] if `return_dict` is True, otherwise a `tuple`. When returning a tuple, the first element is a list with the generated images. ) rrrr8r9rr:rr]FNrrscale)r[rrrrrrr8r9rr:rrr\r]rrcpu)r!)rCr)total)rtimestepencoder_hidden_statespooled_projectionsrrr)rr7r8r9r:latent)r)images)@rUrFrrrrr _pag_scale_pag_adaptive_scalermrnrr,rsrlrgetrrrdo_perturbed_attention_guidance%_prepare_perturbed_attention_guidanceror XLA_AVAILABLEr2r-maxorderrr;rN in_channelsrr^attn_processors_set_pag_attn_processorrC progress_bar enumeraterexpand#_apply_perturbed_attention_guidancerchunkstepbackendsmps is_availablerxlocalspopupdatexm mark_stepr<scaling_factor shift_factordecoderP postprocessmaybe_free_model_hooksset_attn_processorr)1rXr[rrrrrr!rrrrr\rr7r8r9rr:rrrrrrr]rrr{rrtimestep_devicer num_warmup_stepsroriginal_attn_procritlatent_model_inputr noise_prednoise_pred_uncondnoise_pred_text latents_dtypecallback_kwargsrcallback_outputsimages1 r1__call__$StableDiffusion3PAGPipeline.__call__s#DK433d6K6KKI11D4I4II      +//'#9!5*G/Q 3  " .#'=$##5   *VS"9"9J  Jvt$<$<VJ&,,Q/J''?C>Y>Y>eD ' ' + +GT :ko    +//(,(H(H'#9!5*Gnn"7 3!!    " )(  / / FF5t7W7WM$(#M#M$&CTEeEe$  - -!II'=}&MSTUM#(99.KMa-bhi#j  =#O$O); NN/* & &s9~0CnnFZFZ0ZZ\]^!)n $//66BB&& . .            / /!%!1!1!A!A   ( (#'#:#:,0,L,L )    %8  9\!),1>>&+YYyM$$ ((,,## ]]_12#'+/+/! #%%) #264848,-DH,026;?9=BF"' 8< $BF9B #&$'9x<d3ix<S /D(x<S /D( x< d x< Tz x<!x<U d"x<x<tCy4/x<c?T1x<c?T1x< #Tzx<??T%//%::TAx<""T)x< ((4/!x<"!& 1 1D 8#x<$$//$6%x<&(-'8'84'?'x<(4Z)x<*+x<,!%S#X 5-x<.:/x<0'Sz4'784?1x<2-1I3x<4!5x<67x<8"9x<3xr<s 1>_089..O))MM   H %.'+(,"&!% 8*t8* %,,  %8*Cy4 8* K$  8*v^ <"35GI\^f^