3jSrSSKJr SSKrSSKrSSKrSSKrSSKrSSK J r SSK J r SSK JrJr SSKJr SS KJrJrJrJr \"5(aSSKr\R0"\5r"S S \\ 5r\ "S S 55r\ "SS\55r\ "SS\55r\ "SS\55r \ "SS\55r!\ "SS\55r"g)z Adapted from https://github.com/huggingface/transformers/blob/52cb4034ada381fe1ffe8d428a1076e5411a8026/src/transformers/utils/quantization_config.py ) annotationsN) dataclass)Enum)AnyCallable)version) deprecateis_torch_availableis_torchao_versionloggingc(\rSrSrSrSrSrSrSrSr g) QuantizationMethod- bitsandbytesgguftorchaoquantomodeloptN) __name__ __module__ __qualname____firstlineno__BITS_AND_BYTESGGUFTORCHAOQUANTOMODELOPT__static_attributes__rb/home/wildlama/miniconda3/lib/python3.13/site-packages/diffusers/quantizers/quantization_config.pyrr-s#N DG FHr!rcp\rSrSr%SrS\S'/r\SSj5rSSjr SSjr Sr S r SSS jjr S rS rg )QuantizationConfigMixin5z% Mixin class for quantization config r quant_methodc U"S0UD6n/nUR5H4upg[XF5(dM[XFU5 URU5 M6 UHnUR US5 M U(aXC4$U$)an Instantiates a [`QuantizationConfigMixin`] from a Python dictionary of parameters. Args: config_dict (`dict[str, Any]`): Dictionary that will be used to instantiate the configuration object. return_unused_kwargs (`bool`, *optional*, defaults to `False`): Whether or not to return a list of unused keyword arguments. Used for `from_pretrained` method in `PreTrainedModel`. kwargs (`dict[str, Any]`): Additional parameters from which to initialize the configuration object. Returns: [`QuantizationConfigMixin`]: The configuration object instantiated from those parameters. Nr)itemshasattrsetattrappendpop)cls config_dictreturn_unused_kwargskwargsconfig to_removekeyvalues r" from_dict!QuantizationConfigMixin.from_dict>sw$#{#  ,,.JCv##U+  %)C JJsD ! > !Mr!c[USSS9nUR5n[R"USSS9S-nUR U5 SSS5 g!,(df  g=f) a Save this instance to a JSON file. Args: json_file_path (`str` or `os.PathLike`): Path to the JSON file in which this configuration instance's parameters will be saved. use_diff (`bool`, *optional*, defaults to `True`): If set to `True`, only the difference between the config instance and the default `QuantizationConfig()` is serialized to JSON file. wzutf-8)encodingr Tindent sort_keys N)opento_dictjsondumpswrite)selfjson_file_pathwriterr. json_strings r" to_json_file$QuantizationConfigMixin.to_json_file_sL.# 8F,,.K**[dKdRK LL % 9 8 8s ;A AcB[R"UR5$) Serializes this instance to a Python dictionary. Returns: `dict[str, Any]`: Dictionary of all the attributes that make up this configuration instance. )copydeepcopy__dict__rCs r"r?QuantizationConfigMixin.to_dictps }}T]]++r!c## [R"UR5R5H upX4v M g7f)zTallows `dict(obj)` for situations where obj may be a dict or QuantizationConfigMixinN)rKrLrMr()rCattrr4s r"__iter__ QuantizationConfigMixin.__iter__ws0==7==?KD+ @s>AcTURRSUR53$)N ) __class__rto_json_stringrNs r"__repr__ QuantizationConfigMixin.__repr__|s(..))*!D,?,?,A+BCCr!cUSLaUR5nOUR5n[R"USSS9S-$)ar Serializes this instance to a JSON string. Args: use_diff (`bool`, *optional*, defaults to `True`): If set to `True`, only the difference between the config instance and the default `PretrainedConfig()` is serialized to JSON string. Returns: `str`: String containing all the attributes that make up this configuration instance in JSON format. Tr r:r=) to_diff_dictr?r@rA)rCuse_diffr.s r"rW&QuantizationConfigMixin.to_json_strings= t ++-K,,.Kzz+a4@4GGr!c /nUR5H4up4[X5(dM[XU5 URU5 M6 UR5VVs0sHup4X2;dM X4_M nnnU$s snnf)ao Updates attributes of this class instance with attributes from `kwargs` if they match existing attributes, returning all the unused kwargs. Args: kwargs (`dict[str, Any]`): Dictionary of attributes to tentatively update this class. Returns: `dict[str, Any]`: Dictionary containing all the key-value pairs that were not used to update the instance. )r(r)r*r+)rCr0r2r3r4 unused_kwargss r"updateQuantizationConfigMixin.updatesq  ,,.JCt!!5)  %) 7=lln]n H\n ]^s  A8-A8rNF)rDzstr | os.PathLikereturnzdict[str, Any])T)r\boolrdstr)rrrr__doc____annotations___exclude_attributes_at_init classmethodr5rGr?rRrXrWr`r rr!r"r$r$5sH%$"$@&", DH$r!r$c\rSrSrSr/SQrSSjr\S5r\RSSj5r\S5r \ RSS j5r S r S r S r SS jrSrSSjrSrg)BitsAndBytesConfiga This is a wrapper class about all possible attributes and features that you can play with a model that has been loaded using `bitsandbytes`. This replaces `load_in_8bit` or `load_in_4bit` therefore both options are mutually exclusive. Currently only supports `LLM.int8()`, `FP4`, and `NF4` quantization. If more methods are added to `bitsandbytes`, then more arguments will be added to this class. Args: load_in_8bit (`bool`, *optional*, defaults to `False`): This flag is used to enable 8-bit quantization with LLM.int8(). load_in_4bit (`bool`, *optional*, defaults to `False`): This flag is used to enable 4-bit quantization by replacing the Linear layers with FP4/NF4 layers from `bitsandbytes`. llm_int8_threshold (`float`, *optional*, defaults to 6.0): This corresponds to the outlier threshold for outlier detection as described in `LLM.int8() : 8-bit Matrix Multiplication for Transformers at Scale` paper: https://huggingface.co/papers/2208.07339 Any hidden states value that is above this threshold will be considered an outlier and the operation on those values will be done in fp16. Values are usually normally distributed, that is, most values are in the range [-3.5, 3.5], but there are some exceptional systematic outliers that are very differently distributed for large models. These outliers are often in the interval [-60, -6] or [6, 60]. Int8 quantization works well for values of magnitude ~5, but beyond that, there is a significant performance penalty. A good default threshold is 6, but a lower threshold might be needed for more unstable models (small models, fine-tuning). llm_int8_skip_modules (`list[str]`, *optional*): An explicit list of the modules that we do not want to convert in 8-bit. This is useful for models such as Jukebox that has several heads in different places and not necessarily at the last position. For example for `CausalLM` models, the last `lm_head` is typically kept in its original `dtype`. llm_int8_enable_fp32_cpu_offload (`bool`, *optional*, defaults to `False`): This flag is used for advanced use cases and users that are aware of this feature. If you want to split your model in different parts and run some parts in int8 on GPU and some parts in fp32 on CPU, you can use this flag. This is useful for offloading large models such as `google/flan-t5-xxl`. Note that the int8 operations will not be run on CPU. llm_int8_has_fp16_weight (`bool`, *optional*, defaults to `False`): This flag runs LLM.int8() with 16-bit main weights. This is useful for fine-tuning as the weights do not have to be converted back and forth for the backward pass. bnb_4bit_compute_dtype (`torch.dtype` or str, *optional*, defaults to `torch.float32`): This sets the computational type which might be different than the input type. For example, inputs might be fp32, but computation can be set to bf16 for speedups. bnb_4bit_quant_type (`str`, *optional*, defaults to `"fp4"`): This sets the quantization data type in the bnb.nn.Linear4Bit layers. Options are FP4 and NF4 data types which are specified by `fp4` or `nf4`. bnb_4bit_use_double_quant (`bool`, *optional*, defaults to `False`): This flag is used for nested quantization where the quantization constants from the first quantization are quantized again. bnb_4bit_quant_storage (`torch.dtype` or str, *optional*, defaults to `torch.uint8`): This sets the storage type to pack the quanitzed 4-bit prarams. kwargs (`dict[str, Any]`, *optional*): Additional parameters from which to initialize the configuration object. ) _load_in_4bit _load_in_8bitr&Nc  ^[RTlU(aU(a [S5eUTlUTlUTlUTlUTlUTl UTl U Tl Uc[RTlO][U[ 5(a[#[U5TlO2[U[R$5(aUTlO [S5eU c[R&TlOn[U [ 5(a'U S;a [S5e[#[U 5TlO2[U [R$5(aU TlO [S5eU (aW[+U4SjU 55(d=[,R/S[1U R355STR4S 35 TR75 g) NVload_in_4bit and load_in_8bit are both True, but only one can be used at the same timez8bnb_4bit_compute_dtype must be a string or a torch.dtype)float16float32int8uint8float64bfloat16zv`bnb_4bit_quant_storage` must be a valid string (one of 'float16', 'float32', 'int8', 'uint8', 'float64', 'bfloat16') z8bnb_4bit_quant_storage must be a string or a torch.dtypec3@># UHoTR;v M g7fN)ri).0krCs r" .BitsAndBytesConfig.__init__..sTV4#C#CCVszUnused kwargs: z. These kwargs are not used in .)rrr& ValueErrorrornllm_int8_thresholdllm_int8_skip_modules llm_int8_enable_fp32_cpu_offloadllm_int8_has_fp16_weightbnb_4bit_quant_typebnb_4bit_use_double_quanttorchrsbnb_4bit_compute_dtype isinstancerfgetattrdtyperubnb_4bit_quant_storageallloggerwarninglistkeysrV post_init) rC load_in_8bit load_in_4bitrrrrrrrrr0s ` r"__init__BitsAndBytesConfig.__init__s/== Luv v))"4%:"0P-(@%#6 )B& ! )*/--D ' . 4 4*1%9O*PD ' . < <*@D 'WX X ! )*/++D ' . 4 4%.!M+2%9O*PD ' . < <*@D 'WX X #TVTTT NN_T&++--@,AA`aeaoao`ppqr s r!cUR$ry)rnrNs r"rBitsAndBytesConfig.load_in_4bit!!!r!c[U[5(d [S5eUR(aU(a [ S5eXlg)Nload_in_4bit must be a booleanrq)rre TypeErrorrrrnrCr4s r"rr!8%&&<= =   uv v"r!cUR$ry)rorNs r"rBitsAndBytesConfig.load_in_8bit*rr!c[U[5(d [S5eUR(aU(a [ S5eXlg)Nload_in_8bit must be a booleanrq)rrerrrrors r"rr.rr!c[UR[5(d [S5e[UR[5(d [S5e[UR [ 5(d [S5eURb*[UR[5(d [S5e[UR[5(d [S5e[UR[5(d [S5eURb4[UR[R5(d [S5e[UR[5(d [S 5e[UR [5(d [S 5eUR(aW["R$"[&R(R#S 55["R$"S 5:d [+S 5egg)zn Safety checker that arguments are correct - also replaces some NoneType arguments with their default values. rrz"llm_int8_threshold must be a floatNz/llm_int8_skip_modules must be a list of stringsz2llm_int8_enable_fp32_cpu_offload must be a booleanz*llm_int8_has_fp16_weight must be a booleanz*bnb_4bit_compute_dtype must be torch.dtypez$bnb_4bit_quant_type must be a stringz+bnb_4bit_use_double_quant must be a booleanrz0.39.0z[4 bit quantization requires bitsandbytes>=0.39.0 - please upgrade your bitsandbytes version)rrrerrrfloatrrrrrrrrrfrrparse importlibmetadatarrNs r"rBitsAndBytesConfig.post_init7s$++T22<= =$++T22<= =$11599@A A  % % 1*TE_E_ae:f:fMN N$??FFPQ Q$77>>HI I  & & 2:dFaFachcncn;o;oHI I$22C88BC C$88$??IJ J   W]]93E3E3M3Mn3]%^bibobo c & m & r!c@UR=(d UR$)z@ Returns `True` if the model is quantizable, `False` otherwise. )rrrNs r"is_quantizable!BitsAndBytesConfig.is_quantizable\s  5D$5$55r!cUR(agUR(aURS:XagUR(aURS:Xagg)zu This method returns the quantization method used for the model. If the model is not quantizable, it returns `None`. llm_int8fp4nf4N)rrrrNs r"quantization_method&BitsAndBytesConfig.quantization_methodbsE      4#;#;u#D   4#;#;u#Dr!c[R"UR5n[US5R S5SUS'[US5R S5SUS'UR US'UR US'U$)rJrr~rrr)rKrLrMrfsplitrr)rCoutputs r"r?BitsAndBytesConfig.to_dictps t}}-+.v6N/O+P+V+VWZ+[\]+^'(+.v6N/O+P+V+VWZ+[\]+^'(!%!2!2~!%!2!2~ r!cUR5nURRS[R"USSS9S3$NrUr Tr:r=r?rVrr@rArCr.s r"rXBitsAndBytesConfig.__repr__};lln ..))*!DJJ{1X\,]+^^`aar!cUR5n[5R5n0nUR5HupEXRU:wdMXSU'M U$)z Removes all attributes from config which correspond to the default config attributes for better readability and serializes to a Python dictionary. Returns: `dict[str, Any]`: Dictionary of all the attributes that make up this configuration instance, )r?rlr()rCr.default_config_dictserializable_config_dictr3r4s r"r[BitsAndBytesConfig.to_diff_dicts[lln 12::<#% &++-JCC0005-.('r!) rnrorrrrrrrrr&) FFg@NFFNrFN)r4rerc)rrrrrgrirpropertyrsetterrrrrr?rXr[r rr!r"rlrls1f#U").!&#!"'#<|""##""###J6  b(r!rlc&\rSrSrSrSSSjjrSrg)GGUFQuantizationConfigia5This is a config class for GGUF Quantization techniques. Args: compute_dtype: (`torch.dtype`, defaults to `torch.float32`): This sets the computational type which might be different than the input type. For example, inputs might be fp32, but computation can be set to bf16 for speedups. Nc[RUlXlSUlSUlURc[ RUlgg)NT)rrr& compute_dtype pre_quantizedmodules_to_not_convertrrs)rCrs r"rGGUFQuantizationConfig.__init__sE.33*!'+#    %!&D  &r!)rrrr&ry)rz'torch.dtype' | None)rrrrrgrr rr!r"rrs / /r!rcl^\rSrSrSrS S SjjrSrU4Sjr\S Sj5r Sr Sr S r U=r $) TorchAoConfigiaThis is a config class for torchao quantization/sparsity techniques. Args: quant_type (`AOBaseConfig`): An `AOBaseConfig` subclass instance specifying the quantization type. See the [torchao documentation](https://docs.pytorch.org/ao/main/api_ref_quantization.html#inference-apis-for-quantize) for available config classes (e.g. `Int4WeightOnlyConfig`, `Int8WeightOnlyConfig`, `Float8WeightOnlyConfig`, `Float8DynamicActivationFloat8WeightConfig`, etc.). modules_to_not_convert (`list[str]`, *optional*, default to `None`): The list of modules to not quantize, useful for quantizing models that explicitly require to have some modules left in their original precision. Example: ```python from diffusers import FluxTransformer2DModel, TorchAoConfig from torchao.quantization import Int8WeightOnlyConfig quantization_config = TorchAoConfig(Int8WeightOnlyConfig()) transformer = FluxTransformer2DModel.from_pretrained( "black-forest-labs/Flux.1-Dev", subfolder="transformer", quantization_config=quantization_config, torch_dtype=torch.bfloat16, ) ``` c f[RUlXlX lUR 5 gry)rrr& quant_typerr)rCrrr0s r"rTorchAoConfig.__init__s' /66$&<# r!c[SS5(a [S5eSSKJn [ UR U5(d+[ S[UR 5R35eg)N<0.15.0zWTorchAoConfig requires torchao >= 0.15.0. Please upgrade with `pip install -U torchao`.r) AOBaseConfigz1quant_type must be an AOBaseConfig instance, got ) r rtorchao.quantization.quant_apirrrrtyper)rCrs r"rTorchAoConfig.post_inits[ c8 , ,vw w?$//<88OPTUYUdUdPePnPnOopq q9r!c^>[TU]5nSSKJn SU"UR50US'U$)z&Convert configuration to a dictionary.r)config_to_dictdefaultr)superr?torchao.core.configrr)rCdrrVs r"r?TorchAoConfig.to_dicts2 GO  7 %nT__&EF,r!c [SS5(d [S5eUR5nURS5n[ U5S:XaSU;dS5eUSnSS KJn U"U5nU"S SU0UD6$) z'Create configuration from a dictionary.z>=rzCTorchAoConfig requires torchao >= 0.15.0 for construction from dictrrrz8Expected only one key 'default' in quant_type dictionaryr)config_from_dictr)r NotImplementedErrorrKr,lenrr)r-r.r/r0rrs r"r5TorchAoConfig.from_dicts"$11%&kl l!&&(  __\2 :!# Z(? F ? *  9%j1 8j8K88r!cUR$)zBCreate the appropriate quantization method based on configuration.)rrNs r"get_apply_tensor_subclass'TorchAoConfig.get_apply_tensor_subclasss r!cUR5nURRS[R"USSS9S3$rrrs r"rXTorchAoConfig.__repr__ rr!)rr&rry)rz'AOBaseConfig'rlist[str] | NonerdNonerb)rrrrrgrrr?rjr5rrXr __classcell__)rVs@r"rrs^<48 " !1   r 99*bbr!rc6\rSrSrSrSSSjjrSrSrg) QuantoConfigia] This is a wrapper class about all possible attributes and features that you can play with a model that has been loaded using `quanto`. Args: weights_dtype (`str`, *optional*, defaults to `"int8"`): The target dtype for the weights after quantization. Supported values are ("float8","int8","int4","int2") modules_to_not_convert (`list`, *optional*, default to `None`): The list of modules to not quantize, useful for quantizing models that explicitly require to have some modules left in their original precision (e.g. Whisper encoder, Llava encoder, Mixtral gate layers). Nc Sn[SSU5 [RUlXlX lUR 5 g)NzB`QuantoConfig` is deprecated and will be removed in version 1.0.0.rz1.0.0)r rrr& weights_dtyperr)rCrrr0deprecation_messages r"rQuantoConfig.__init__s< c.'+>?.55*&<# r!cb/SQnURU;a[SUSUR35eg)z+ Safety checker that arguments are correct )float8rtint4int2zOnly support weights in z but found N)rr)rCaccepted_weightss r"rQuantoConfig.post_init*sA>   %5 578H7IUYUgUgThij j 6r!)rr&r)rtN)rrfrr)rrrrrgrrr rr!r"rrs, $37  !1 kr!rc\rSrSrSrSSSSSS.rSSS.rSSS jjrSSS jjrSS jr SS jr Sr g )NVIDIAModelOptConfigi3a9This is a config class to use nvidia modelopt for quantization. Args: quant_type (`str`): The type of quantization we want to use, following is how to use: **weightquant_activationquant ==> FP8_FP8** In the above example we have use FP8 for both weight and activation quantization. Following are the all the options: - FP8 - INT8 - INT4 - NF4 - NVFP4 modules_to_not_convert (`list[str]`, *optional*, default to `None`): The list of modules to not quantize, useful for quantizing models that explicitly require to have some weight_only (`bool`, *optional*, default to `False`): If set to `True`, the quantization will be applied only to the weights of the model. channel_quantize (`int`, *optional*, default to `None`): The channel quantization axis, useful for quantizing models across different axes. block_quantize (`int`, *optional*, default to `None`): The block size, useful to further quantize each channel/axes into blocks. scale_channel_quantize (`int`, *optional*, default to `None`): The scale channel quantization axis, useful for quantizing calculated scale across different axes. scale_block_quantize (`int`, *optional*, default to `None`): The scale block size, useful for quantizing each scale channel/axes into blocks. algorithm (`str`, *optional*, default to `"max"`): The algorithm to use for quantization, currently only supports `"max"`. forward_loop (`Callable`, *optional*, default to `None`): The forward loop function to use for calibration during quantization. modelopt_config (`dict`, *optional*, default to `None`): The modelopt config, useful for passing custom configs to modelopt. disable_conv_quantization (`bool`, *optional*, default to `False`): If set to `True`, the quantization will be disabled for convolutional layers. kwargs (`dict[str, Any]`, *optional*): Additional parameters which are to be used for calibration. )r)r r)FP8INT8INT4NF4NVFP4)rrNc  [RUlURU5 X lX0lX@lXPlSU0UlXl X`l Xpl U (dUR5OU Ul Xlg)Nmethod)rrr&_normalize_quant_typer weight_onlychannel_quantizeblock_quantize calib_cfg forward_loopscale_channel_quantizescale_block_quantizeget_config_from_quant_typemodelopt_configdisable_conv_quantization) rCrrrrrrr algorithmrr r r0s r"rNVIDIAModelOptConfig.__init__esv/77 "":.&<#& 0, i )&<#$8!HWt>>@]l)B&r!chSSKJn [U5S:XaSUS3n[R"U5 gg)Nr)_PATCHED_CLASSESzNot z weights in modelopt format. This might cause unreliable behavior.Please make sure to run the following code before loading/saving model weights: from modelopt.torch.opt import enable_huggingface_checkpointing enable_huggingface_checkpointing() )&modelopt.torch.opt.plugins.huggingfacerrwarningswarn)rC operationr warning_msgs r"check_model_patching)NVIDIAModelOptConfig.check_model_patchings<K  A %yk";;  MM+ & &r!cURS5nUSn[U5S:aUSOSn[U5S:a[RSUS35 SnSnOaU[R ;a[RS US 35 SnUb/U[R ;a[RS US 35 SnX4bSU-OS -Ulg)av Validates and normalizes the quantization type string. Splits the quant_type into weight and activation components, verifies them against supported types, and replaces unsupported values with safe defaults. Args: quant_type (str): The input quantization type string (e.g., 'FP8_INT8'). Returns: str: A valid quantization type string (e.g., 'FP8_INT8' or 'FP8'). _rrNr zQuantization type z. is not supported. Picking FP8_INT8 as defaultrzWeight Quantization type z) is not supported. Picking FP8 as defaultzActivation Quantization type z* is not supported. Picking INT8 as default)rrrrrquanttype_to_numbitsr)rCrpartsw_typeact_types r"r*NVIDIAModelOptConfig._normalize_quant_types  %q"5zA~584 u:> NN/ |;ij kFH1FFF!:6(Bklm#8L8a8a(a!>xjHrst 6JC(NPRSr!cSSKJs Jn SS00SS00000S.URREUR S.nUSnUR (a#UHnS U;dM X4(aMSX4S'M URRS 5nUSn[U5S :aUS RS S 5OSnUHcnXARR;dMSX4;dM(US:XaUb[RUX4S'MK[RUX4S'Me URbLURb?URUR0US S'URURSS0USS'O9URb,URUS S'URUSS'SUSS'URbUR bU[R";aAUS SR%[R"UURUR 0S.5 U(aUU[R";aAUSSR%[R"UURUR 0S.5 U$)z, Get the config from the quantization type. rN fake_quantFenable)*weight_quantizer*input_quantizerz*output_quantizerz*q_bmm_quantizerz*k_bmm_quantizerz*v_bmm_quantizerz*softmax_quantizer) quant_cfgr r$r"rrArr#num_bits block_sizesrdynamicaxis) scale_bitsscale_block_sizes)modelopt.torch.quantizationr quantizationr1_default_disabled_quantizer_cfgrrrrrreplacerrrrrrquanttype_to_scalingbitsr`)rCmtq BASE_CONFIGr$r{rrrs r"r /NVIDIAModelOptConfig.get_config_from_quant_types 21'3E%:$&&.%6$&$&$&&( **<<    ,   &a/ -2IL*%%c*q03E Q58##C,DA BBBxW`WcGc**+3G3\3\]e3f Z0+?+T+TU[+\ Z(     *t/D/D/P=A=R=RTXTgTg 'T:Ar!r)#rg __future__rrKimportlib.metadatarr@osr dataclassesrenumrtypingrr packagingrutilsr r r r r get_loggerrrrfrr$rlrrrrrr!r"r@s$ #  ! NN   H %d oo od l(0l( l(^ /4/ /, [b+[b [b| !k*!k !kH {2{ {r!