1 Release published by 1 person

• v0.34.0 Diffusers 0.34.0: New Image and Video Models, Better torch.compile Support, and more

  published Jun 24, 2025

103 Pull requests merged by 30 people

• [chore] deprecate blip controlnet pipeline.

  #11877 merged Jul 7, 2025

• Modular custom config object serialization

  #11868 merged Jul 5, 2025

• Modular Guider ConfigMixin

  #11862 merged Jul 5, 2025

• Modular PAG Guider

  #11860 merged Jul 4, 2025

• Fix Wan AccVideo/CausVid fuse_lora

  #11856 merged Jul 4, 2025

• FIX set_lora_device when target layers differ

  #11844 merged Jul 4, 2025

• [benchmarks] overhaul benchmarks

  #11565 merged Jul 4, 2025

• [docs] Add a note of _keep_in_fp32_modules

  #11851 merged Jul 2, 2025

• [Flux Kontext] Support Fal Kontext LoRA

  #11823 merged Jul 2, 2025

• update: FluxKontextInpaintPipeline support

  #11820 merged Jul 2, 2025

• [lora] tests for exclude_modules with Wan VACE

  #11843 merged Jul 2, 2025

• [From Single File] support from_single_file method for WanVACE3DTransformer

  #11807 merged Jul 2, 2025

• fix norm not training in train_control_lora_flux.py

  #11832 merged Jul 2, 2025

• [docs] Deprecated pipelines

  #11838 merged Jul 2, 2025

• [docs] Batch generation

  #11841 merged Jul 2, 2025

• Use real-valued instead of complex tensors in Wan2.1 RoPE

  #11649 merged Jul 1, 2025

• [docs] fix single_file example.

  #11847 merged Jul 1, 2025

• [single file] Cosmos

  #11801 merged Jul 1, 2025

• [tests] Fix failing float16 cuda tests

  #11835 merged Jul 1, 2025

• reset deterministic in tearDownClass

  #11785 merged Jul 1, 2025

• [tests] add test for hotswapping + compilation on resolution changes

  #11825 merged Jul 1, 2025

• Remove print statement in SCM Scheduler

  #11836 merged Jun 30, 2025

• ENH: Improve speed of function expanding LoRA scales

  #11834 merged Jun 30, 2025

• [lora]feat: use exclude modules to loraconfig.

  #11806 merged Jun 30, 2025

• [lora] fix: lora unloading behvaiour

  #11822 merged Jun 28, 2025

• Support dynamically loading/unloading loras with group offloading

  #11804 merged Jun 27, 2025

• TorchAO compile + offloading tests

  #11697 merged Jun 27, 2025

• [CI] disable onnx, mps, flax from the CI

  #11803 merged Jun 27, 2025

• remove syncs before denoising in Kontext

  #11818 merged Jun 27, 2025

• Kontext fixes

  #11815 merged Jun 26, 2025

• Kontext training

  #11813 merged Jun 26, 2025

• Flux Kontext

  #11812 merged Jun 26, 2025

• adjust tolerance criteria for test_float16_inference in unit test

  #11809 merged Jun 26, 2025

• [tests] add a test on torch compile for varied resolutions

  #11776 merged Jun 26, 2025

• [rfc][compile] compile method for DiffusionPipeline

  #11705 merged Jun 26, 2025

• Follow up for Group Offload to Disk

  #11760 merged Jun 26, 2025

• [chore] post release v0.34.0

  #11800 merged Jun 26, 2025

• fix deprecation in lora after 0.34.0 release

  #11802 merged Jun 25, 2025

• adjust to get CI test cases passed on XPU

  #11759 merged Jun 25, 2025

• [tests] skip instead of returning.

  #11793 merged Jun 25, 2025

• guard omnigen processor.

  #11799 merged Jun 24, 2025

• [tests] Fix group offloading and layerwise casting test interaction

  #11796 merged Jun 24, 2025

• [chore] raise as early as possible in group offloading

  #11792 merged Jun 24, 2025

• [tests] Fix HunyuanVideo Framepack device tests

  #11789 merged Jun 24, 2025

• [lora] only remove hooks that we add back

  #11768 merged Jun 24, 2025

• [docs] minor cleanups in the lora docs.

  #11770 merged Jun 24, 2025

• Add --lora_alpha and metadata handling to train_dreambooth_lora_sana.py

  #11744 merged Jun 23, 2025

• [Wan] Fix mask padding in Wan VACE pipeline.

  #11778 merged Jun 23, 2025

• [CI] Skip ONNX Upscale tests

  #11774 merged Jun 23, 2025

• [tests] properly skip tests instead of return

  #11771 merged Jun 23, 2025

• enable cpu offloading of new pipelines on XPU & use device agnostic empty to make pipelines work on XPU

  #11671 merged Jun 23, 2025

• enable deterministic in bnb 4 bit tests

  #11738 merged Jun 23, 2025

• Fix dimensionalities in apply_rotary_emb functions' comments

  #11717 merged Jun 21, 2025

• [docs] LoRA scale scheduling

  #11727 merged Jun 20, 2025

• [docs] device_map

  #11711 merged Jun 20, 2025

• [docs] Quantization + torch.compile + offloading

  #11703 merged Jun 20, 2025

• Fix failing cpu offload test for LTX Latent Upscale

  #11755 merged Jun 20, 2025

• fix invalid component handling behaviour in PipelineQuantizationConfig

  #11750 merged Jun 20, 2025

• [modular diffusers] more refactor

  #11235 merged Jun 20, 2025

• Update Chroma Docs

  #11753 merged Jun 19, 2025

• make group offloading work with disk/nvme transfers

  #11682 merged Jun 19, 2025

• Fix HiDream pipeline test module

  #11754 merged Jun 19, 2025

• [CI] Fix SANA tests

  #11756 merged Jun 19, 2025

• [CI] Fix WAN VACE tests

  #11757 merged Jun 19, 2025

• [LoRA] refactor lora loading at the model-level

  #11719 merged Jun 19, 2025

• Bump urllib3 from 2.2.3 to 2.5.0 in /examples/server

  #11748 merged Jun 19, 2025

• save_pretrained for serializing config.

  #11603 merged Jun 19, 2025

• Add missing HiDream license

  #11747 merged Jun 19, 2025

• Update more licenses to 2025

  #11746 merged Jun 19, 2025

• [Quantizers] add is_compileable property to quantizers.

  #11736 merged Jun 19, 2025

• Chroma Follow Up

  #11725 merged Jun 18, 2025

• [chore] change to 2025 licensing for remaining

  #11741 merged Jun 18, 2025

• [tests] device_map tests for all models.

  #11708 merged Jun 18, 2025

• [training] add ds support to lora hidream

  #11737 merged Jun 18, 2025

• ⚡️ Speed up method AutoencoderKLWan.clear_cache by 886%

  #11665 merged Jun 18, 2025

• [LoRA training] update metadata use for lora alpha + README

  #11723 merged Jun 17, 2025

• Support more Wan loras (VACE)

  #11726 merged Jun 17, 2025

• Add Pruna optimization framework documentation

  #11688 merged Jun 16, 2025

• Fix misleading comment

  #11722 merged Jun 16, 2025

• [training] show how metadata stuff should be incorporated in training scripts.

  #11707 merged Jun 16, 2025

• [LoRA ]fix flux lora loader when return_metadata is true for non-diffusers

  #11716 merged Jun 16, 2025

• Chroma Pipeline

  #11698 merged Jun 14, 2025

• Cosmos Predict2

  #11695 merged Jun 13, 2025

• [LoRA] parse metadata from LoRA and save metadata

  #11324 merged Jun 13, 2025

• Support Wan AccVideo lora

  #11704 merged Jun 13, 2025

• [docs] mention fp8 benefits on supported hardware.

  #11699 merged Jun 13, 2025

• swap out token for style bot.

  #11701 merged Jun 13, 2025

• [docs] add compilation bits to the bitsandbytes docs.

  #11693 merged Jun 12, 2025

• Apply Occam's Razor in position embedding calculation

  #11562 merged Jun 11, 2025

• Avoid DtoH sync from access of nonzero() item in scheduler

  #11696 merged Jun 11, 2025

• Set _torch_version to N/A if torch is disabled.

  #11645 merged Jun 11, 2025

• Improve Wan docstrings

  #11689 merged Jun 11, 2025

• [tests] model-level device_map clarifications

  #11681 merged Jun 11, 2025

• [tests] tests for compilation + quantization (bnb)

  #11672 merged Jun 11, 2025

• enable torchao test cases on XPU and switch to device agnostic APIs for test cases

  #11654 merged Jun 11, 2025

• [Wan] Standardize vae.encode() sampling mode in WanVideoToVideoPipeline

  #11639 merged Jun 11, 2025

• [LoRA] support Flux Control LoRA with bnb 8bit.

  #11655 merged Jun 11, 2025

• Allow remote code repo names to contain "."

  #11652 merged Jun 10, 2025

• Update pipeline_flux_inpaint.py to fix padding_mask_crop returning only the inpainted area

  #11658 merged Jun 10, 2025

• Add community class StableDiffusionXL_T5Pipeline

  #11626 merged Jun 9, 2025

• Introduce DeprecatedPipelineMixin to simplify pipeline deprecation process

  #11596 merged Jun 9, 2025

• [tests] Fix how compiler mixin classes are used

  #11680 merged Jun 9, 2025

• fixed axes_dims_rope init (huggingface#11641)

  #11678 merged Jun 8, 2025

38 Pull requests opened by 21 people

• Fix EDM DPM Solver Test and Enhance Test Coverage

  #11679 opened Jun 8, 2025

• [GGUF] feat: support loading diffusers format gguf checkpoints.

  #11684 opened Jun 10, 2025

• [WIP] Refactor Attention Modules

  #11685 opened Jun 10, 2025

• Bump requests from 2.32.3 to 2.32.4 in /examples/server

  #11686 opened Jun 10, 2025

• fix "Expected all tensors to be on the same device, but found at least two devices" error

  #11690 opened Jun 11, 2025

• [docs] quant_kwargs

  #11712 opened Jun 14, 2025

• Add MAGI-1: Autoregressive Video Generation at Scale

  #11713 opened Jun 14, 2025

• [tests] annotate compilation test classes with bnb

  #11715 opened Jun 15, 2025

• [Docker] update doc builder dockerfile to include quant libs.

  #11728 opened Jun 17, 2025

• Add cond + uncond batching to Chroma

  #11729 opened Jun 17, 2025

• Fix lora alpha and metadata handling

  #11739 opened Jun 18, 2025

• [tests] add compile + offload tests for GGUF.

  #11740 opened Jun 18, 2025

• [research_projects] add shortened flux training script with quantization

  #11743 opened Jun 18, 2025

• [wip] enable FastPersist to (potentially) speed up disk saving in group offloading

  #11758 opened Jun 19, 2025

• Avoid creating tensor in CosmosAttnProcessor2_0 (#11761)

  #11763 opened Jun 21, 2025

• Add `--lora_alpha` and metadata handling for `train_dreambooth_lora_hidream`

  #11765 opened Jun 21, 2025

• Fix unique memory address when doing group-offloading with disk

  #11767 opened Jun 22, 2025

• [ci] unify how adhoc installs are done in CI.

  #11773 opened Jun 23, 2025

• [training] add an offload utility that can be used as a context manager.

  #11775 opened Jun 23, 2025

• [LoRA] add lora_alpha to sana README

  #11780 opened Jun 23, 2025

• Add --lora_alpha and metadata handling to train_dreambooth_lora_sd3.py

  #11783 opened Jun 23, 2025

• [CI] Fix big GPU test marker

  #11786 opened Jun 24, 2025

• [wip] Higgs quantization

  #11797 opened Jun 24, 2025

• Flux: pass joint_attention_kwargs when using gradient_checkpointing

  #11814 opened Jun 26, 2025

• [ci] enable hotswapping tests on our nightly CI.

  #11826 opened Jun 28, 2025

• [Refactor] TensorRTStableDiffusionPipeline - Remove StableDiffusionReferencePipeline inheriting from DiffusionPipeline

  #11829 opened Jun 29, 2025

• [docs] torch.compile blog post

  #11837 opened Jun 30, 2025

• [tests] unbloat `tests/lora/utils.py`

  #11845 opened Jul 1, 2025

• [docs] LoRA metadata

  #11848 opened Jul 1, 2025

• Propose to refactor output normalization in several transformers

  #11850 opened Jul 2, 2025

• [docs] fix references in flux pipelines.

  #11857 opened Jul 4, 2025

• [training ] add Kontext i2i training

  #11858 opened Jul 4, 2025

• [fix] VaeImageProcessor.resize method can take list as input

  #11863 opened Jul 4, 2025

• Add cuda kernel support for GGUF inference

  #11869 opened Jul 5, 2025

• Add STORK scheduler

  #11871 opened Jul 6, 2025

• [fix] fix for prior preservation and mixed precision sampling

  #11873 opened Jul 6, 2025

• add kontext into auto-pipeline

  #11875 opened Jul 7, 2025

• [quant] QoL improvements for pipeline-level quant config

  #11876 opened Jul 7, 2025

47 Issues closed by 17 people

• Multi-ID customization with CogvideoX 5B

  #10462 closed Jul 5, 2025

• When using gradient accumulation, optimizer.step() runs every single step. Is it as expected？

  #10855 closed Jul 5, 2025

• Time series diffusion

  #10886 closed Jul 5, 2025

• train_dreambooth_lora_flux crash on batch size greater than 1

  #10994 closed Jul 5, 2025

• support SageAttention

  #11121 closed Jul 5, 2025

• How to multi-GPU WAN inference

  #11339 closed Jul 5, 2025

• convert_hunyuan_video_to_diffusers.py not working - invalid load key, '\xd8'.

  #11370 closed Jul 5, 2025

• Bug: set adapters when model is compiled

  #11766 closed Jul 5, 2025

• support for MeanFlow

  #11781 closed Jul 5, 2025

• [FR] Please support WanVACETransformer3DModel.from_single_file

  #11630 closed Jul 5, 2025

• set_lora_device reports KeyError even though lora is loaded

  #11833 closed Jul 4, 2025

• Conflix size when load controlnet+IP-Adapter when load stable-diffusion-xl-base-1.0

  #11853 closed Jul 3, 2025

• framepack pipeline return pure black output when using torch.float16

  #11830 closed Jul 3, 2025

• Single File and GGUF support of Cosmos-Predict2

  #11798 closed Jul 1, 2025

• flux_control support flux kontext dev

  #11817 closed Jul 1, 2025

• set_adapters performance degrades with the number of inactive adapters

  #11816 closed Jun 30, 2025

• FLUX Kontext. Transformers error: safetensors_rust.SafetensorError: Error while deserializing header: MetadataIncompleteBuffer

  #11827 closed Jun 29, 2025

• can't load lora weights after apply group offloading

  #11791 closed Jun 27, 2025

• flux fill fp8 load failed

  #10244 closed Jun 27, 2025

• Flash Attention

  #11805 closed Jun 25, 2025

• Add `--lora_alpha` and metadata handling in training scripts follow up

  #11730 closed Jun 23, 2025

• Incorrect mask padding in Wan VACE pipeline

  #11779 closed Jun 23, 2025

• How to convert the hidream lora trained by diffusers to a format that comfyui can load?

  #11396 closed Jun 23, 2025

• The docs notebooks are not updated since Jul 26, 2023

  #10900 closed Jun 23, 2025

• [performance] investigating FluxPipeline for recompilations on resolution changes

  #11360 closed Jun 23, 2025

• Reporting a vulnerability

  #11714 closed Jun 20, 2025

• SanaPipeline crash on load

  #11721 closed Jun 20, 2025

• Pipeline Quantization silent ignore of components not present in pipeline

  #11745 closed Jun 20, 2025

• AnimateDiffPipeline: Tensor Mismatch with num_frames > motion_max_seq_length (chunking failure)

  #11615 closed Jun 18, 2025

• Chroma fails to mask attention in the transformer

  #11724 closed Jun 18, 2025

• FLUX.1-dev Controlnet Training Script for Custom Dataset.

  #11734 closed Jun 18, 2025

• Regression? load_lora_weights ValueError: not enough values to unpack (expected 3, got 2)

  #11732 closed Jun 17, 2025

• FluxFillPipeline Support load IP Adapter.

  #11561 closed Jun 17, 2025

• CausVid LoRA key error when loading with diffusers

  #11709 closed Jun 17, 2025

• [DOCS] Add `pruna` as optimization framework

  #11687 closed Jun 16, 2025

• torch.compile errors on vae.encode

  #10937 closed Jun 14, 2025

• Support Chroma - Flux based model with architecture changes

  #11010 closed Jun 14, 2025

• AccVid LoRA key error when loading with diffusers

  #11702 closed Jun 13, 2025

• Convert VAE from latent-diffusion to diffusers

  #11694 closed Jun 11, 2025

• <spam>

  #11692 closed Jun 11, 2025

• <spam>

  #11691 closed Jun 11, 2025

• Add support for ConsisID

  #10100 closed Jun 10, 2025

• HunyuanVideo with IP2V

  #10485 closed Jun 10, 2025

• Docs for HunyuanVideo LoRA?

  #10796 closed Jun 10, 2025

• Need to handle v0.33.0 deprecations

  #10895 closed Jun 10, 2025

• What's the difference between 'example/train_text_to_image_lora.py' and 'example/research_projects/lora/train_text_to_image_lora.py' ?

  #11542 closed Jun 10, 2025

• [BUG] [CleanCode] Tuple[int] = (16, 56, 56) in FluxTransformer2DModel

  #11641 closed Jun 8, 2025

30 Issues opened by 28 people

• WanVACETransformer3DModel with GGUF not working for 1.3B model

  #11878 opened Jul 7, 2025

• [lora] `exclude_modules` won't consider the modules names in `target_modules`

  #11874 opened Jul 7, 2025

• CPU Offloading Inefficiency: UNet/Transformer Repeatedly Moved Between CPU/GPU During Loops

  #11872 opened Jul 6, 2025

• Question about control sd3

  #11867 opened Jul 5, 2025

• [Docs] Visualizing the codebase for diffusers

  #11866 opened Jul 5, 2025

• AttributeError: type object 'CosmosTransformer3DModel' has no attribute 'from_single_file'

  #11865 opened Jul 5, 2025

• AutoencoderDC.encode fails with torch.compile(fullgraph=True) - "name 'torch' is not defined"

  #11864 opened Jul 5, 2025

• ControlnetUnion not working correctly

  #11861 opened Jul 4, 2025

• STORK Scheduler

  #11859 opened Jul 4, 2025

• Can not load fusionx_lora into original wan2.1-14b

  #11849 opened Jul 2, 2025

• Finetuning DreamBoot Inpainting

  #11846 opened Jul 1, 2025

• finetune flux kontext on custom dataset

  #11840 opened Jul 1, 2025

• Try load gguf for FluxKontextPipeline but invalid input size

  #11839 opened Jul 1, 2025

• StableDiffusionXLControlNetUnionInpaintPipeline requires a float for controlnet_conditioning_scale & control_guidance_end

  #11828 opened Jun 29, 2025

• Run FluxKontextPipeline with multi-images input, But An error occurred

  #11824 opened Jun 28, 2025

• WanVACEPipeline: Passing a list of prompts is not yet supported. This may be supported in the future.

  #11821 opened Jun 27, 2025

• LoRA Training Script for Image Editing with FLUX Kontext

  #11819 opened Jun 27, 2025

• Bug: lora trained with train_dreambooth_lora_sdxl_advanced.py, the weight cannot be loaded

  #11790 opened Jun 24, 2025

• device_map can not use xla device such as tpu

  #11788 opened Jun 24, 2025

• Bug diffuser

  #11784 opened Jun 23, 2025

• Some question about training lora for HiDream-I1

  #11777 opened Jun 23, 2025

• RepositoryNotFoundError on inpainting demo

  #11764 opened Jun 21, 2025

• Avoid creating tensor in CosmosAttnProcessor2_0

  #11761 opened Jun 20, 2025

• Failed to load lora in int8 mode

  #11752 opened Jun 19, 2025

• FlowMatchEulerDiscreteScheduler Timesteps have Floating Point Errors

  #11749 opened Jun 19, 2025

• Optional 'prompt_2' parameter in encode_prompt method is not optional

  #11742 opened Jun 18, 2025

• ControlNet LoRA support more broadly across models

  #11733 opened Jun 17, 2025

• add T5 lora train and infer for FluxPipeline

  #11720 opened Jun 16, 2025

• [FR] Please support MagCache: Fast Video Generation with Magnitude-Aware Cache

  #11710 opened Jun 13, 2025

• Add pruna integration for loading model through diffusers.from_pretrained / pipeline

  #11700 opened Jun 12, 2025

53 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• Add SkyReels V2: Infinite-Length Film Generative Model

  #11518 commented on Jul 5, 2025 • 117 new comments

• The Modular Diffusers

  #9672 commented on Jul 7, 2025 • 83 new comments

• Add Finegrained FP8

  #11647 commented on Jun 23, 2025 • 49 new comments

• Add VidTok AutoEncoders

  #11261 commented on Jul 2, 2025 • 13 new comments

• Attention Dispatcher

  #11368 commented on Jun 8, 2025 • 8 new comments

• First Block Cache

  #11180 commented on Jul 7, 2025 • 3 new comments

• Add FluxPAGPipeline with support for PAG

  #11510 commented on Jun 11, 2025 • 2 new comments

• Fix wrong param types, docs, and handles noise=None in scale_noise of FlowMatching schedulers

  #11669 commented on Jun 13, 2025 • 2 new comments

• [Refactor] Initialize class members inside init method

  #11450 commented on Jun 16, 2025 • 1 new comment

• Add SUPIR Upscaler

  #7219 commented on Jun 10, 2025 • 0 new comments

• Dreambooth finetune FLUX dev CLIPTextModel

  #10925 commented on Jul 7, 2025 • 0 new comments

• Error in init from pretrained for LTXConditionPipeline

  #11644 commented on Jul 7, 2025 • 0 new comments

• [Fix Issue #1197 since 2022] Support pre-trained openai/guided-diffusion (ADM) with minimal code change

  #6730 commented on Jul 4, 2025 • 0 new comments

• Tensor-parallel like FeedForward to lower memory requirements

  #10623 commented on Jul 2, 2025 • 0 new comments

• Support for control-lora

  #10686 commented on Jul 5, 2025 • 0 new comments

• Add gradient checkpointing support for AutoencoderKLWan

  #11105 commented on Jul 5, 2025 • 0 new comments

• [Quantization] Add TRT-ModelOpt as a Backend

  #11173 commented on Jun 29, 2025 • 0 new comments

• ⚡️ Speed up method `Kandinsky3ConditionalGroupNorm.forward` by 7%

  #11667 commented on Jun 16, 2025 • 0 new comments

• [Feature] MultiControlNet support for SD3Impainting

  #11251 commented on Jun 16, 2025 • 0 new comments

• [WIP] [LoRA] support omi hidream lora.

  #11660 commented on Jun 10, 2025 • 0 new comments

• Add basic implementation of AuraFlowImg2ImgPipeline

  #11340 commented on Jun 11, 2025 • 0 new comments

• Update pipeline_flux_fill.py

  #11622 commented on Jun 14, 2025 • 0 new comments

• [LoRA training] add aspect ratio bucketing

  #11438 commented on Jun 25, 2025 • 0 new comments

• [NewModel] Adds SEVA ( Stable Virtual Camera) model

  #11440 commented on Jun 16, 2025 • 0 new comments

• [quant] add __repr__ for better printing of configs.

  #11452 commented on Jun 18, 2025 • 0 new comments

• [torch.compile] Make HiDream torch.compile ready

  #11477 commented on Jun 17, 2025 • 0 new comments

• Added PhotoDoodle Pipeline

  #11621 commented on Jun 12, 2025 • 0 new comments

• The density_for_timestep_sampling and loss_weighting for SD3 Training！！！

  #9056 commented on Jun 10, 2025 • 0 new comments

• max_shard_size

  #11650 commented on Jun 10, 2025 • 0 new comments

• OMI Format Compatibility

  #11631 commented on Jun 10, 2025 • 0 new comments

• torch.compile can't be used with groupoffloading on hunyuanvideo_frampack

  #11584 commented on Jun 11, 2025 • 0 new comments

• HunyuanVideoImageToVideoPipeline memory leak

  #11676 commented on Jun 11, 2025 • 0 new comments

• [Performance] Issue on *SanaLinearAttnProcessor2_0 family. 1.06X speedup can be reached with a simple change.

  #11499 commented on Jun 12, 2025 • 0 new comments

• Can't load flux-fill-lora with FluxControl

  #11651 commented on Jun 12, 2025 • 0 new comments

• Request support for MAGI-1

  #11519 commented on Jun 12, 2025 • 0 new comments

• possibly to avoid `from_single_file` loading in fp32 to save RAM

  #10679 commented on Jun 13, 2025 • 0 new comments

• self.attn ValueError: not enough values to unpack (expected 2, got 1)

  #10066 commented on Jun 19, 2025 • 0 new comments

• how to load lora weight with fp8 transfomer model?

  #11648 commented on Jun 19, 2025 • 0 new comments

• prompt_embeds_scale in FluxPriorReduxPipeline seems to have no effect.

  #11642 commented on Jun 21, 2025 • 0 new comments

• StableDiffusionXLControlNetInpaintPipeline unable to use padding_mask_crop with multiple controlnets

  #10573 commented on Jun 23, 2025 • 0 new comments

• Sage Attention for diffuser library

  #11168 commented on Jun 23, 2025 • 0 new comments

• Run FLUX-controlnet zero3 training failed: 'weight' must be 2-D

  #10969 commented on Jun 24, 2025 • 0 new comments

• Support zero-3 for FLUX training

  #10743 commented on Jun 24, 2025 • 0 new comments

• [FR] Please support ref image and multiple control videos in Wan VACE

  #11674 commented on Jun 25, 2025 • 0 new comments

• Since last update of transformers I can't import diffusers modules

  #11343 commented on Jun 26, 2025 • 0 new comments

• Make `FlaxLMSDiscreteScheduler` jittable

  #2180 commented on Jun 28, 2025 • 0 new comments

• CogvideoX VAE decoder consumes significantly more memory in the latest version

  #10091 commented on Jun 30, 2025 • 0 new comments

• set_adapters not support for compiled model

  #11408 commented on Jul 5, 2025 • 0 new comments

• Multi-gpus Context Parallel training support?

  #11020 commented on Jul 5, 2025 • 0 new comments

• WanImageToVideoPipeline - swap out a limited number of blocks

  #10999 commented on Jul 5, 2025 • 0 new comments

• Comments bug in FluxPriorReduxPipeline

  #10457 commented on Jul 5, 2025 • 0 new comments

• `StableAudioDiTModel` throws `[rank0]: RuntimeError: DDP...` on `accelerator.prepare` when training on 4 GPUs

  #10160 commented on Jul 5, 2025 • 0 new comments

• Error when I'm trying to train a Flux lora with train_dreambooth_lora_flux_advanced

  #11489 commented on Jul 6, 2025 • 0 new comments