huggingface/diffusers

Issues

• `lora_scale` has no effect when loading with Flux

  #9525 opened in 3 hours by cshowley
  0

• Error Mapping on sd3, sdxl and upcoming flux controlnet training scripts

  #9523 opened in 3 hours by Night1099
  0

• UNetMotionModel.dtype is really expensive to call, is it possible to cache it during inference?

  #9520 opened in 3 hours by xiang9156
  2

• Can we allow making everything on gpu/cuda for scheduler?

  #9485 opened 5 days ago by xiang9156
  10

• self.scheduler.add_noise did not add any noise for the given latent

  #9519 opened in 3 hours by LonglongaaaGo
  0

• parameters `joint_attention_kwargs` doesn't be passed to FLUX's transformers model

  #9516 opened 21 hours ago by HorizonWind2004
  1

• The `JointAttnProcessor2_0` in the `SD3Transformer2DModel` does not include `RMSNorm`.

  #9515 opened 21 hours ago by HuiZhang0812
  0

• Flux-dev - Unable to load LoRA weights after fp8 quantisation

  #9514 opened 21 hours ago by kishlaykumar1995
  1

• train_dreambooth_lora_flux validation RuntimeError: Input type (float) and bias type (c10::BFloat16) should be the same

  #9476 opened 5 days ago by squewel
  2

• Multi-controlnet batching for `StableDiffusionXLControlNetInpaintPipeline`

  #9511 opened 21 hours ago by kdubovikov
  0

• train_dreambooth_lora_flux with prodigy and train_text_encoder causes IndexError: list index out of range

  #9464 opened 6 days ago by squewel
  4

• [Schedulers] Add SGMUniform

  #9416 opened 21 hours ago by asomoza
  12

• [Schedulers] Analysis of `simple`, `exponential`, `polyexponential` and `beta`

  #9490 opened 4 days ago by hlky
  2

• SD3ControlNetModel forward function error

  #9496 opened 21 hours ago by pibbo88
  3

• Dreambooth Flux training failed on saving a checkpoint

  #9500 opened 2 days ago by kopyl
  6

• Dreambooth Flux training error: RuntimeError: mat2 must be a matrix, got 1-D tensor

  #9497 opened 2 days ago by kopyl
  3

• AnimateDiff SparseCtrl RGB does not work as expected

  #9508 opened 21 hours ago by a-r-r-o-w
  0

• Can't upscale anything. Keep getting "KeyError: 'middle_block_out.0.weight'" error/

  #9505 opened 21 hours ago by evanerichards
  2

• Remove redundant comparison inside the diffusion loop of stable video diffusion pipeline

  #9425 opened 13 days ago by dianyo
  6

• Dreambooth Flux training does not save a model for around 10-15 minutes

  #9501 opened 2 days ago by kopyl
  4

• FLUX dreambooth train on multigpu with deepspeed

  #9484 opened 5 days ago by zhangvia
  1

• Confusion about `FrozenDict` in `configuration_utils.py`

  #9503 opened 2 days ago by townwish4git
  0

• ValueError: Multiple file extensions found at ./models/coreml-stable-diffusion-v1-4_original_packages.Cannot infer resource type from contents.

  #9471 opened 6 days ago by LukeLIN-web
  4

• SDXL PAG with IPAdapter is not working.

  #9495 opened 2 days ago by akshatd007
  2

• Lumina pipeline fails to generate any image

  #9488 opened 2 days ago by zhuole1025
  4

• FluxPipeline - Multi-GPU Issue - When you define transformer= you get "Expected all tensors to be on the same device"

  #9450 opened 8 days ago by CrackerHax
  8

• Add FLUX image-to-image to `AutoPipelineForImage2Image`

  #9492 opened 2 days ago by apolinario
  1

• Problem with FluxInpaintPipeline when doing a replace.

  #9486 opened 3 days ago by ukaprch
  8

• Add GGUF loader for FluxTransformer2DModel

  #9487 opened 4 days ago by vladmandic
  3

• Prompt scheduling in Diffusers like A1111

  #9470 opened 6 days ago by linhbeige
  2

• StableDiffusionInpaintPipeline changes areas that I didn't mask out.

  #9479 opened 5 days ago by arceus-jia
  0

• [BUG] 'GatheredParameters' object is not callable

  #9477 opened 5 days ago by LinB203
  1

• Formulation of reverse diffusion process in DDPM ( 1 - alpha_prod_t = beta_prod_t assumption)

  #9474 opened 5 days ago by jadevaibhav
  3

• Inconsistent results from Flux Model when loaded differently

  #9439 opened 10 days ago by emil-malina
  9

• Unable to Access 'runwayml/stable-diffusion-inpainting' Model

  #9421 opened 5 days ago by HaoyueBaiZJU
  3

• CogVideoX I2V: Missing guard-rail on num_frames

  #9467 opened 6 days ago by tin2tin
  3

• Flux inference og

  #9461 opened 6 days ago by echosyy
  3

• AttributeError: 'tuple' object has no attribute 'shape' while using IP-Adapter with StableDiffusionControlNetInpaintPipeline

  #9448 opened 6 days ago by satvik-pyxer
  3

• CogVideoX Image2Video model not found

  #9454 opened 6 days ago by rolux
  2

• Enable assigning a list of SD3ControlNetModel to StableDiffusion3ControlNetPipeline

  #9460 opened 7 days ago by avrech
  1

• [BUG] SNR gamma in v_prediction

  #9451 opened 8 days ago by LinB203
  1

• model_name in quicktour page results in a 404 error

  #9438 opened 8 days ago by vinayakkgarg
  3

• SD3 Error: Cannot instantiate this tokenizer from a slow version.

  #9443 opened 9 days ago by Scorpinaus
  1

• Suggestion for speeding up `index_for_timestep` by removing sequential `nonzero()` calls in samplers

  #9417 opened 14 days ago by ethanweber
  4

• The transformer model saved after FLUX+Hyper-SD lora cannot be loaded

  #9420 opened 12 days ago by peki12345
  3

• AttributeError: 'NoneType' object has no attribute 'get'

  #9430 opened 12 days ago by molo32
  0

• d3_dreambooth_lora_16gb.ipynb broken for me with latest, ok with released version

  #9424 opened 13 days ago by 3a1b2c3
  7

• A strange thing happened when I wrote my own code to train Cotrolnet_sdxl, as soon as I did the first backpropagation, noise_pred became nan.

  #9422 opened 13 days ago by Li-Zn-H
  5

• add new parameter module of CogVideoXTransformer3DModel, raise error

  #9415 opened 13 days ago by trouble-maker007
  2

• CogvideoX-5b adds freeze-frames in start and end of clip

  #9413 opened 14 days ago by tin2tin
  1