huggingface/diffusers

Issues

• Getting CUDA out of memory error even with Colab A100 high RAM

  #11014 opened 11 hours ago by gurselnaziroglu
  0

• Support wan2.1 video model?

  #11008 opened 2 days ago by kexul
  1

• Support Chroma - Flux based model with architecture changes

  #11010 opened 2 days ago by nitinmukesh
  4

• WanImageToVideoPipeline - swap out a limited number of blocks

  #10999 opened 3 days ago by spezialspezial
  6

• Broken video output with Wan 2.1 I2V pipeline + quantized transformer

  #11006 opened 2 days ago by rolux
  4

• pipeline_wan_i2v.py: minor discrepancy between arg default and docstring

  #11005 opened 2 days ago by rolux
  0

• Flux Lora Unloading Issue Discussion

  #11003 opened 3 days ago by CyberVy
  2

• Flux Controlnet Lora Fails To Load When Transformers Are Quantized.

  #10989 opened 4 days ago by CyberVy
  12

• Any chance class members like self._interrupt could be defined in __init__ across pipelines?

  #11002 opened 3 days ago by spezialspezial
  0

• Diffusers Transformer Pipeline Produces ComplexDouble Tensors on MPS, Causing Conversion Error

  #10986 opened 4 days ago by mozzipa
  1

• Loading WanTransformer3DModel using torch_dtype=torch.bfloat16 keeps some parameters as float32

  #10992 opened 3 days ago by spezialspezial
  5

• WanImageToVideoPipeline.__call__ missing shift argument

  #11000 opened 3 days ago by spezialspezial
  2

• train_dreambooth_lora_flux crash on batch size greater than 1

  #10994 opened 3 days ago by PluginBOXone
  2

• f-divergence

  #10993 opened 3 days ago by manmeet3591
  4

• WAN 2.1 T2V Unable to Quantize the model based on using QuantoConfig

  #10988 opened 4 days ago by ukaprch
  2

• Spatio-temporal diffusion models

  #10987 opened 4 days ago by moghadas76
  0

• Another comfy-ui compatible FLUX1.dev LoRA fails to load

  #10954 opened 6 days ago by AmericanPresidentJimmyCarter
  11

• Loading LoRA weights fails for OneTrainer Flux LoRAs

  #10972 opened 4 days ago by spezialspezial
  2

• diffusers/utils/import_utils.py No module named 'triton.ops'

  #10940 opened 5 days ago by ZanePoe
  6

• Cannot copy out of meta tensor; no data! with CogView4Pipeline

  #10958 opened 6 days ago by nitinmukesh
  5

• Cogview4 pipeline not accepting prompt embeds, due to shape issues .

  #10962 opened 5 days ago by Vargol
  1

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

  #10969 opened 5 days ago by alien-0119
  0

• Integrate LaVie (IJCV 2024) and Cinemo (CVPR 2025) to diffusers

  #10967 opened 5 days ago by maxin-cn
  0

• Accelerate Inference Section of Doc Broken

  #10950 opened 5 days ago by AbhinavGopal
  2

• cannot import name 'AutoencoderKLWan' from 'diffusers'

  #10963 opened 6 days ago by spawner1145
  3

• should use 'permute' replace 'transpose'

  #10953 opened 6 days ago by nku-shengzheliu
  1

• 【SDXLLongPromptWeightingPipeline load lora】ValueError: Adapter name(s) {'toy'} not in the list of present adapters: {'default_0'}.

  #10952 opened 6 days ago by kolar1988
  0

• ValueError: Attempting to unscale FP16 gradients.

  #10910 opened 7 days ago by Messimanda
  3

• torch.compile errors on vae.encode

  #10937 opened 8 days ago by Luciennnnnnn
  1

• parameter missing in Documentation and mat cant be multipled

  #10935 opened 8 days ago by Rts-Wu
  0

• Loading dataset from disk where it exists

  #10913 opened 11 days ago by surya-narayanan
  1

• error load IP-adapter xl plus version

  #10933 opened 9 days ago by john09282922
  0

• HunyuanVideo pipe.transformer.compile(): torch._dynamo hit config.recompile_limit (8)

  #10932 opened 9 days ago by eppaneamd
  0

• Is lumina-2.0 script correct?

  #10917 opened 11 days ago by Riko0
  3

• OSError: Error no file named diffusion_pytorch_model.fp16.bin found in directory /local/data/RealVisXL_V5.0/unet

  #10928 opened 10 days ago by kolar1988
  3

• Dreambooth finetune FLUX dev CLIPTextModel

  #10925 opened 10 days ago by Wuyiche
  0

• need Wan2.1 of diffusers version

  #10903 opened 13 days ago by zhaoyun0071
  1

• [DDIMInverseScheduler] `inf` values at first iteration when `set_alpha_to_one=True` and `prediction_type="sample"`

  #10920 opened 11 days ago by andreabosisio
  0

• Model getting offloaded to CPU without user's intention

  #10914 opened 11 days ago by janzd
  0

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

  #10900 opened 13 days ago by albertvillanova
  4

• CLIP Score Evaluation without Pre-processing.

  #10904 opened 12 days ago by e-delaney
  0

• Whether lohaconfig is supported in the convert_state_dict_to_diffusers method

  #10899 opened 13 days ago by llm8047
  1

• HunyuanVIdeo in diffusers use negative_prompt but generate wrong video

  #10901 opened 13 days ago by philipwan
  0

• Support SkyReels-A1 for expressive portrait animation.

  #10897 opened 13 days ago by qiudi0127
  0

• Support SkyReels-A1 for expressive portrait animation.

  #10896 opened 13 days ago by qiudi0127
  0

• Need to handle v0.33.0 deprecations

  #10895 opened 13 days ago by a-r-r-o-w
  0

• Adding support for Datasets in the StableDiffusionPipeline

  #10892 opened 13 days ago by sashavor
  2

• [Feature request] CogvideoX Controlnet integration for 5B / 2B

  #10893 opened 13 days ago by nitinmukesh
  0

• Suggestion to add TEs

  #10890 opened 14 days ago by janekm
  1

• Internal type conversions breaks DCAE on bfloat16

  #10891 opened 14 days ago by SwayStar123
  1