bi+ (ddlZddlmZmZmZmZmZmZddlZ ddl Z ddl m Z m Z mZmZmZmZddlmZmZddlmZmZmZmZddlmZmZddlmZdd lm Z m!Z!m"Z"m#Z#m$Z$m%Z%dd l&m'Z'd d l(m)Z)d dl*m+Z+e!r ddl,m-cm.Z/dZ0ndZ0e"jbe2Z3dZ4gdZ5 d#de6de6de7de7fdZ8 d$dee6deee9e jtfdeee6deee7fdZ; d%de jxdee jzde9fd Z>Gd!d"e)eeeeZ?y)&N)AnyCallableDictListOptionalUnion)CLIPImageProcessor CLIPTextModel CLIPTokenizerCLIPVisionModelWithProjectionT5EncoderModelT5TokenizerFast)PipelineImageInputVaeImageProcessor)FluxIPAdapterMixinFluxLoraLoaderMixinFromSingleFileMixinTextualInversionLoaderMixin) AutoencoderKLFluxTransformer2DModel)FlowMatchEulerDiscreteScheduler)USE_PEFT_BACKENDis_torch_xla_availableloggingreplace_example_docstringscale_lora_layersunscale_lora_layers) randn_tensor)DiffusionPipeline)FluxPipelineOutputTFa Examples: ```py >>> import torch >>> from diffusers import FluxKontextPipeline >>> from diffusers.utils import load_image >>> pipe = FluxKontextPipeline.from_pretrained( ... "black-forest-labs/FLUX.1-Kontext-dev", torch_dtype=torch.bfloat16 ... ) >>> pipe.to("cuda") >>> image = load_image( ... "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/diffusers/yarn-art-pikachu.png" ... ).convert("RGB") >>> prompt = "Make Pikachu hold a sign that says 'Black Forest Labs is awesome', yarn art style, detailed, vibrant colors" >>> image = pipe( ... image=image, ... prompt=prompt, ... guidance_scale=2.5, ... generator=torch.Generator().manual_seed(42), ... ).images[0] >>> image.save("output.png") ``` )) )))p) 0)@)p)P)r4)r3r2)r1r0)r/r.)r-r,)r+r*)r)r()r'r&)r%r$ base_seq_len max_seq_len base_shift max_shiftc<||z ||z z }|||zz }||z|z}|SN) image_seq_lenr5r6r7r8mbmus i/home/cdr/jupyterlab/.venv/lib/python3.12/site-packages/diffusers/pipelines/flux/pipeline_flux_kontext.pycalculate_shiftrAgs; Z K,$>?AQ%%A  Q B Inum_inference_stepsdevice timestepssigmasc | | td|dttj|jj j v}|std|jd|jd ||d||j}t|}||fS|dttj|jj j v}|std|jd|jd ||d||j}t|}||fS|j|fd |i||j}||fS) a Calls the scheduler's `set_timesteps` method and retrieves timesteps from the scheduler after the call. Handles custom timesteps. Any kwargs will be supplied to `scheduler.set_timesteps`. Args: scheduler (`SchedulerMixin`): The scheduler to get timesteps from. num_inference_steps (`int`): The number of diffusion steps used when generating samples with a pre-trained model. If used, `timesteps` must be `None`. device (`str` or `torch.device`, *optional*): The device to which the timesteps should be moved to. If `None`, the timesteps are not moved. timesteps (`List[int]`, *optional*): Custom timesteps used to override the timestep spacing strategy of the scheduler. If `timesteps` is passed, `num_inference_steps` and `sigmas` must be `None`. sigmas (`List[float]`, *optional*): Custom sigmas used to override the timestep spacing strategy of the scheduler. If `sigmas` is passed, `num_inference_steps` and `timesteps` must be `None`. Returns: `Tuple[torch.Tensor, int]`: A tuple where the first element is the timestep schedule from the scheduler and the second element is the number of inference steps. zYOnly one of `timesteps` or `sigmas` can be passed. Please choose one to set custom valuesrEzThe current scheduler class zx's `set_timesteps` does not support custom timestep schedules. Please check whether you are using the correct scheduler.)rErDrFzv's `set_timesteps` does not support custom sigmas schedules. Please check whether you are using the correct scheduler.)rFrDrDr;) ValueErrorsetinspect signature set_timesteps parameterskeys __class__rElen) schedulerrCrDrErFkwargsaccepts_timesteps accept_sigmass r@retrieve_timestepsrUus>!3tuu'3w/@/@AXAX/Y/d/d/i/i/k+ll .y/B/B.CDab  M)FMfM'' !)n ) ))   C(9(9):Q:Q(R(](](b(b(d$ee .y/B/B.CD_`  GvfGG'' !)n ) ))  3MFMfM'' ) ))rBencoder_output generator sample_modect|dr |dk(r|jj|St|dr|dk(r|jjSt|dr |jSt d)N latent_distsampleargmaxlatentsz3Could not access latents of provided encoder_output)hasattrrZr[moder]AttributeError)rVrWrXs r@retrieve_latentsrast~}-+2I))00;;  /K84K))..00  +%%%RSSrBc?0eZdZdZdZddgZddgZ dNded ed e d e d e d e de dedeffd Z dOdeeeefdededeej.deej0f dZ dPdeeeefdedeej.fdZ dQdeeeefdeeeeefdeej.dedeej6deej6dedeefdZdZdZ dRdZ e!d Z"e!d!Z#e!d"Z$d#ejJd$ejLfd%Z'd&Z(d'Z)d(Z*d)Z+ dNd#eejJd*ed+ed,ed-edej0dej.d$eeejLeejLfdeejJfd.Z,e-d/Z.e-d0Z/e-d1Z0e-d2Z1e-d3Z2ejfe4e5dddddd4ddd5dd6dddddddddddd7d8dddgdd9d8fd#ee6deeeefdeeeeefd:eeeefd;eeeeefdeeed?edeed$eeejLeejLfdeej6deej6deej6d@ee6dAeeejJdBee6dCeeejJdDeej6dEeej6dFeedGe7dHee8ee9fdIee:eee8gdfdJeededKedLe7ftext_encoder_2->image_encoder->transformer->vae image_encoderfeature_extractorr] prompt_embedsNrQvae text_encoder tokenizertext_encoder_2 tokenizer_2 transformerc t ||j|||||||||  t|ddr/dt |j j jdz znd|_t|ddr |j j jnd|_ t|jdz|_ t|dr"|j|jjnd |_d |_y) N) rgrhrjrirkrlrQrdrergr r")vae_scale_factorriM)super__init__register_modulesgetattrrPrgconfigblock_out_channelsrplatent_channelsrimage_processorr^rimodel_max_lengthtokenizer_max_lengthdefault_sample_size) selfrQrgrhrirjrkrlrdrerOs r@rtzFluxKontextPipeline.__init__s  %)##'/  W^^bdikoVpc$((//*L*L&MPQ&Q RvwCJ$PUW[B\txx>>bd0$BWBWZ[B[\/6t[/IdnnNhDNN + +np !$' rBr"promptnum_images_per_promptmax_sequence_lengthrDdtypec |xs |j}|xs|jj}t|tr|gn|}t |}t|t r|j||j}|j|d|dddd}|j}|j|ddj} | jd|jdk\rbtj|| sL|jj| dd|jd z df} tj!d |d | |j#|j%|d d } |j"j}| j%||} | j\} } } | j'd |d } | j)||z| d} | S)N max_lengthTFpt)paddingr truncation return_lengthreturn_overflowing_tokensreturn_tensorslongestrrr"zXThe following part of your input was truncated because `max_sequence_length` is set to tokens: output_hidden_statesrrrD)_execution_devicerhr isinstancestrrPrmaybe_convert_promptrk input_idsshapetorchequal batch_decoder|loggerwarningrjtorepeatview)r~rrrrDr batch_size text_inputstext_input_idsuntruncated_ids removed_textrf_seq_lens r@_get_t5_prompt_embedsz)FluxKontextPipeline._get_t5_prompt_embedss14110**00'4&&[ d7 8..vt7G7GHF&&  *&+' %..**69UY*Zdd   $(<(>66q$JcJcfgJgjlJlGl7mnL NN--.i ~G )).*;*;F*CZ_)` &33 %((t/@/@/F/Fv(V &,,Q0EF %**:8M+MrR rBprompt_2pooled_prompt_embeds lora_scalec l|xs |j}|gt|trW||_|jt rt |j||jt rt |j|t|tr|gn|}|D|xs|}t|tr|gn|}|j|||}|j||||}|j,t|trt rt|j||j,t|trt rt|j||j|jjn|jj} tj|j ddj#|| } ||| fS)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 device: (`torch.device`): torch device num_images_per_prompt (`int`): number of images that should be generated per prompt 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. 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. lora_scale (`float`, *optional*): A lora scale that will be applied to all LoRA layers of the text encoder if LoRA layers are loaded. )rrDr)rrrrDr"rrDr)rrr _lora_scalerhrrrjrrrrrrlrzerosrr) r~rrrDrrfrrrrtext_idss r@ encode_promptz!FluxKontextPipeline.encode_promptfs@1411  !j7J&K)D   ,1A!$"3"3Z@"".3C!$"5"5zB'4&&  )6H%/#%>zHH$(#?#?&;$@$ !66&;$7 7M    ($ 349I#D$5$5zB    *$ 349I#D$7$7D+/+<+<+H!!''dN^N^NdNd;;}2215q9< ? @ '&rBc r|jdzzdk7s|jdzzdk7r,tjdjdzd|d|d| Ltfd| Ds8t dj d | D cgc]} | j vs| c} ||t d |d |d ||t d |d |d | | t d|7t |ts't |tst dt||7t |ts't |tst dt|||t d|d|d ||t d|d|d | | t d| | t d| | dkDrt d| yycc} w)Nr rz-`height` and `width` have to be divisible by z but are  and z(. Dimensions will be resized accordinglyc3:K|]}|jvywr:)_callback_tensor_inputs).0kr~s r@ z3FluxKontextPipeline.check_inputs..s#F 23A-- -F sz2`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`: 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 z'Cannot forward both `negative_prompt`: z and `negative_prompt_embeds`: z)Cannot forward both `negative_prompt_2`: 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`.rz8`max_sequence_length` cannot be greater than 512 but is ) rprrallrHrrrrtype)r~rrheightwidthnegative_promptnegative_prompt_2rfnegative_prompt_embedsrnegative_pooled_prompt_embeds"callback_on_step_end_tensor_inputsrrs` r@ check_inputsz FluxKontextPipeline.check_inputss T**Q. /1 4AVAVYZAZ8[_`8` NN?@U@UXY@Y?ZZcdjckkpqvpwx`a  . 9#F 7YF C DTEaEaDbbn|^pHvwbckokGkGbGpqpHoIJ   -";08N}o^00  !m&?28*RSTVZSZ>[#[ #3F#;ell5>QRVXYRY>Z#Z RbRhRhO 57O+33 "%: :>rBc|j|||dzd|dzd}|jdddddd}|j||dz|dzz|dz}|S)Nr rr"r)rpermuter)r]rnum_channels_latentsrrs r@ _pack_latentsz!FluxKontextPipeline._pack_latents;sl,,z+?1aQVZ[Q[]^_//!Q1a3//*v{uz.JL`cdLderBc|j\}}}dt||dzzz}dt||dzzz}|j||dz|dz|dzdd}|jdddddd}|j ||dz||}|S)Nr rrrr"r)rintrrr)r]rrrpr num_patcheschannelss r@_unpack_latentsz#FluxKontextPipeline._unpack_latentsDs-4MM) Kc&k&6&:;<SZ$4q$89:,,z6Q; HPQMSTVWX//!Q1a3//*h5.A65QrBrrWc t|trkt|jdDcgc]2}t |j j |||dz||d4}}tj|d}n't |j j ||d}||j jjz |j jjz}|Scc}w)Nrr"r\)rWrXr) rrrangerrargencoderrrw shift_factorscaling_factor)r~rrWi image_latentss r@_encode_vae_imagez%FluxKontextPipeline._encode_vae_imageUs i &u{{1~.!q1q51A!BiXYlhpqM"IIm;M,TXX__U-CyfnoM&)E)EEIgIgg s7C)c8|jjy)z Enable sliced VAE decoding. When this option is enabled, the VAE will split the input tensor in slices to compute decoding in several steps. This is useful to save some memory and allow larger batch sizes. N)rgenable_slicingr~s r@enable_vae_slicingz&FluxKontextPipeline.enable_vae_slicingd !rBc8|jjy)z Disable sliced VAE decoding. If `enable_vae_slicing` was previously enabled, this method will go back to computing decoding in one step. N)rgdisable_slicingrs r@disable_vae_slicingz'FluxKontextPipeline.disable_vae_slicingls   "rBc8|jjy)a Enable tiled VAE decoding. When this option is enabled, the VAE will split the input tensor into tiles to compute decoding and encoding in several steps. This is useful for saving a large amount of memory and to allow processing larger images. N)rg enable_tilingrs r@enable_vae_tilingz%FluxKontextPipeline.enable_vae_tilingts  rBc8|jjy)z Disable tiled VAE decoding. If `enable_vae_tiling` was previously enabled, this method will go back to computing decoding in one step. N)rgdisable_tilingrs r@disable_vae_tilingz&FluxKontextPipeline.disable_vae_tiling}rrBrrrrc :t|tr)t||k7rtdt|d|ddt ||j dzzz}dt ||j dzzz}||||f} dx} } |?|j ||}|jd|jk7r|j||} n|} || jdkDrC|| jdzdk(r.|| jdz} tj| g| zd } n^|| jdkDr4|| jdzdk7rtd | jdd |d tj| gd } | jdd\}}|j| ||||} |j||dz|dz||} d| d <|j||dz|dz||}| %t| |||} |j| ||||} n| j ||} | | || fS)Nz/You have passed a list of generators of length z+, but requested an effective batch size of z@. Make sure the batch size matches the length of the generators.r rr")rrWrrz'Cannot duplicate `image` of batch size z to z text prompts.).r)rWrDr)rrrPrHrrprrryrrrrrr)r~rrrrrrrDrWr]rr image_idsadditional_image_per_promptimage_latent_heightimage_latent_width latent_idss r@prepare_latentsz#FluxKontextPipeline.prepare_latentss i &3y>Z+GA#i.AQR&<'gi c&kd&;&;a&?@ASZD$9$9A$=>?165A$((  HHF%H8E{{1~!5!55 $ 6 6Ui 6 X % M//22zMDWDWXYDZ7Z^_7_.8MQ>QRS>T=UUYZdYeest!& =/q A 6C6I6I!"6M 3 !3 ..z+?ATVhM66/146HA6MvW\I!"If 33J! UVWZY_afg ?"5IfTYZG((*>RTZ\abGjjejdz}-|jA||"| z|-|||jB|#| | \}}.}/}0|0t'jD|/|0gd }/| tGjHdd | z | n| } |jd }1tK|1|jLj<j#dd|jLj<j#dd|jLj<j#dd|jLj<j#dd}2tO|jL| |#| |2\}3} tQt|3| |jLjRzz d }4t|3|_*|j:j<jVrGt'jXd g| |#t&jZ}5|5j]|jd }5nd}5||Q|O|MtGj^||dftFj`}|g|j:jbjdz}nT|R|P||LtGj^||dftFj`}|g|j:jbjdz}|j i|_d}6d}7|||jg|||#|"| z}6|||jg|||#|"| z}7|jLjid |jk| 5}8tm|3D]I\}9}:|jnr|:|_|6|6|jd<|};|.t'jD||.gd };|:j]|jd jq|jB}<|j;|;||>ddd|j+d f}>|>||=|>z zz}=|jB}?|jLjs|=|:|d d }|jB|?k7r9t&jtjvjyr|jq|?}|Hi}@|D]}At{|A@|A<|||9|:@}B|Bj}d!|}|Bj}d"|}|9t|3d z k(s'|9d z|4kDr/|9d z|jLjRzd k(r|8jts6tjL dddd|_|d#k(r|}n|j||||j}||jj<jz |jj<jz}|jj|d d }|j.j||$}|j|s|fSt|%S#1swYxYw)&a+" Function invoked when calling the pipeline for generation. Args: image (`torch.Tensor`, `PIL.Image.Image`, `np.ndarray`, `List[torch.Tensor]`, `List[PIL.Image.Image]`, or `List[np.ndarray]`): `Image`, numpy array or tensor representing an image batch to be used as the starting point. For both numpy array and pytorch tensor, the expected value range is between `[0, 1]` If it's a tensor or a list or tensors, the expected shape should be `(B, C, H, W)` or `(C, H, W)`. If it is a numpy array or a list of arrays, the expected shape should be `(B, H, W, C)` or `(H, W, C)` It can also accept image latents as `image`, but if passing latents directly it is not encoded again. 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. 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 `true_cfg_scale` is not greater 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. true_cfg_scale (`float`, *optional*, defaults to 1.0): When > 1.0 and a provided `negative_prompt`, enables true classifier-free guidance. 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 3.5): Embedded guidance scale is enabled by setting `guidance_scale` > 1. Higher `guidance_scale` encourages a model to generate images more aligned with prompt at the expense of lower image quality. Guidance-distilled models approximates true classifier-free guidance for `guidance_scale` > 1. Refer to the [paper](https://huggingface.co/papers/2210.03142) to learn more. 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 ge 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. 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. ip_adapter_image: (`PipelineImageInput`, *optional*): Optional image input to work with IP Adapters. ip_adapter_image_embeds (`List[torch.Tensor]`, *optional*): Pre-generated image embeddings for IP-Adapter. It should be a list of length same as number of IP-adapters. Each element should be a tensor of shape `(batch_size, num_images, emb_dim)`. If not provided, embeddings are computed from the `ip_adapter_image` input argument. negative_ip_adapter_image: (`PipelineImageInput`, *optional*): Optional image input to work with IP Adapters. negative_ip_adapter_image_embeds (`List[torch.Tensor]`, *optional*): Pre-generated image embeddings for IP-Adapter. It should be a list of length same as number of IP-adapters. Each element should be a tensor of shape `(batch_size, num_images, emb_dim)`. If not provided, embeddings are computed from the `ip_adapter_image` 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. 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.flux.FluxPipelineOutput`] 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 512): Maximum sequence length to use with the `prompt`. max_area (`int`, defaults to `1024 ** 2`): The maximum area of the generated image in pixels. The height and width will be adjusted to fit this area while maintaining the aspect ratio. Examples: Returns: [`~pipelines.flux.FluxPipelineOutput`] or `tuple`: [`~pipelines.flux.FluxPipelineOutput`] if `return_dict` is True, otherwise a `tuple`. When returning a tuple, the first element is a list with the generated images. ?r z6Generation `height` and `width` have been adjusted to rz to fit the model requirements.)rrrfrrrrrNFr"rscale)rrrfrrDrrrc3LK|]\}}t||z z ||fywr:)abs)rwh aspect_ratios r@rz/FluxKontextPipeline.__call__..s/3:>!QSA-.153s!$rrrbase_image_seq_lenmax_image_seq_lenr7r8ffffff?)rFr?rr)r)totalri) hidden_statestimestepguidancepooled_projectionsencoder_hidden_statestxt_idsimg_idsrr!)r!r]rflatent)r )images)Kr}rproundrrrr rrrrrrrPrrrgetrrrsizeryrzget_default_height_widthminPREFERRED_KONTEXT_RESOLUTIONSresize preprocessrlrw in_channelsr rrnplinspacerArQrUmaxorderrguidance_embedsfullfloat32expandruint8rrrset_begin_index progress_bar enumeraterrstepbackendsmps is_availablelocalspopupdate XLA_AVAILABLExm mark_steprrgrrdecode postprocessmaybe_free_model_hooksr#)Dr~rrrrrrrrrCrFr rrWr]rfrrrrrrrr r!rr"rrr#r$original_heightoriginal_width multiple_ofrrDrhas_neg_prompt do_true_cfgrnegative_text_idsimg image_height image_widthrrrr rr<r?rEnum_warmup_stepsr5rnegative_image_embedsrOrtlatent_model_inputr4 noise_predneg_noise_pred latents_dtypecallback_kwargsrcallback_outputsr,sD @r@__call__zFluxKontextPipeline.__call__s `K433d6K6KKI11D4I4II*0%v~ x,.367</C78++a/ ${2;&4 _ $(? NNHPUV[U\\{|      +/'#9!5*G/Q 3   .'=$!%  *VS"9J  Jvt$<VJ&,,Q/J''?C>Y>Y>eD ' ' + +GT :ko )4 "$ . \3PX\3\ %q(;^   '!5"7 3!      ""&*4%B&;$7%#   &-!  j &E%**UV-[_[o[oJo(5%(5C(,(<(<(U(UVY(Z %L+&5L/23B_30,; &4{BK';6DL((//|[QE((33E<UE $//66BBaG8<8L8L  . .         9 5 I  J #:BJTZSaS!&9"9;NOgm a(   NN ! ! % %&:C @ NN ! ! % %&94 @ NN ! ! % %lC 8 NN ! ! % %k4 8  *< NN   * & &s9~0CdnnFZFZ0ZZ\]^!)n    " " 2 2zz1#~fEMMZHw}}Q'78HH  (,C,O % -2R2Z(*%1C288(T %)B(CdFVFVFgFgFwFw(w %&+B+J % 15U5a!xx(:"((K  01D4D4D4U4U4e4ee   & & .+-D ( $  '+B+N?? '22 L % 04T4`$($H$H)022 % ! &&q)   %8  9C #\!),B #1>>)*&+NZD001JK%," ,).G]3KQR)S&88GMM!$4588G!--"4%_%';*7$&+/+F+F %.   (+BEI%* ;?KO9B #&!?V0*+V0c49n%V05d3i01 V0 sDI~. V0 $E#tCy.$9: V0V0 V0}V0!V0e%V0V0 (}V0E%//43H"HIJV0%++,V0   1 12!V0"'u'8'89#V0$##56%V0&"*$u||*#?)V0*+34 3E*F+V0,!)):): ;-V0.(00A0A'B/V00c]1V023V04!)c3h 85V06'xc40@$0F'GH7V08-1I9V0:!;V0<=V0>?V03V0rBrc)r.r0r&r1)NNNN)Nr[)@rJtypingrrrrrrnumpyrEr transformersr r r r r rrzrrloadersrrrrmodelsrr schedulersrutilsrrrrrrutils.torch_utilsrpipeline_utilsr!pipeline_outputr#torch_xla.core.xla_modelcore xla_modelrYrX get_loggerrqrr}rArrxrArrDrUrrzrarcr;rBr@rso== Epp;9../))MM   H %4!.       *.15%)$( 8*!#8* U3 ,- .8*S " 8* T%[ ! 8*zck TLL T-5eoo-F T\_ Tp0 p0rB