New Model Request Reply

Workflow runs for New Model Request Reply

• 🚀 [Universal Fix] One-line installer for all Qwen-Image 2.1 & Diffusers bugs

  skipped:
  New Model Request Reply#184:Issue#15102openedbyMiddlesomonitor
  Oct 22, 20266s
  More actions for workflow run 184

• [modular] ConditionalPipelineBlocks should declare its own trigger inputs

  skipped:
  New Model Request Reply#183:Issue#15101openedbyyiyixuxu
  Oct 22, 20262s
  More actions for workflow run 183

• [modular] conditional block triggers should be workflow inputs

  skipped:
  New Model Request Reply#182:Issue#15100openedbyyiyixuxu
  Oct 22, 2026Skipped
  More actions for workflow run 182

• load_video returns mixed-mode frames for GIFs, breaking preprocess_video and AnimateDiff video-to-video

  skipped:
  New Model Request Reply#181:Issue#15098openedbynettanvirdev
  Oct 21, 202612s
  More actions for workflow run 181

• DEIS coefficient precision causes increasing error as inference steps increase

  skipped:
  New Model Request Reply#180:Issue#15090openedbyRobertboy18
  Oct 21, 20269s
  More actions for workflow run 180

• lr_scheduler step bug in examples code

  skipped:
  New Model Request Reply#179:Issue#15087openedbyweixiaohao
  Oct 20, 2026Skipped
  More actions for workflow run 179

• Error "cannot import name 'FqnToConfig' from 'torchao.quantization'" when importing the diffusers library on google colab

  skipped:
  New Model Request Reply#178:Issue#15085openedbyWiNE-iNEFF
  Oct 19, 20261s
  More actions for workflow run 178

• Qwen-Image 2.1 Turbo: VAE tiling introduces vertical streaks at 2048x2048 on NVIDIA/CUDA

  skipped:
  New Model Request Reply#177:Issue#15083openedbyReligious-J
  Oct 19, 20261s
  More actions for workflow run 177

• Follow-up to #14737: drop output_type="latent" from modular decode blocks

  skipped:
  New Model Request Reply#176:Issue#15071openedbyyiyixuxu
  Oct 17, 20262s
  More actions for workflow run 176

• FLUX.3: match the released BFL reference image sizing

  skipped:
  New Model Request Reply#175:Issue#15068openedbyyiyixuxu
  Oct 16, 20261s
  More actions for workflow run 175

• Warn when a copy of pipeline.blocks is modified

  skipped:
  New Model Request Reply#174:Issue#15067openedbyyiyixuxu
  Oct 16, 20261s
  More actions for workflow run 174

• Quick survey: your experience as a PyTorch Ecosystem maintainer

  skipped:
  New Model Request Reply#173:Issue#15065openedbymichellearoth
  Oct 16, 20267s
  More actions for workflow run 173

• Third-order DPM++ multistep coefficients fail a quadratic forcing test

  skipped:
  New Model Request Reply#172:Issue#15056openedbyRobertboy18
  Oct 15, 20261s
  More actions for workflow run 172

• [LTX-2] Modular pipelines: let callers skip the audio decode for video-only outputs

  skipped:
  New Model Request Reply#171:Issue#15044openedbychristopher5106
  Oct 15, 20267s
  More actions for workflow run 171

• Warn when add_noise runs before set_timesteps

  skipped:
  New Model Request Reply#170:Issue#15043openedbyTejasS1233
  Oct 15, 20261s
  More actions for workflow run 170

• [Wan2.2 VAE, Qwen-Image-2.1 VAE] Redundant x.clone() in residual up/down blocks keeps an extra activation alive

  skipped:
  New Model Request Reply#169:Issue#15040openedbyluca-888
  Oct 15, 20261s
  More actions for workflow run 169

• LTX2 HDR reference preprocessing applies an unintended stride-aligned resize

  skipped:
  New Model Request Reply#168:Issue#15037openedbyShenrui-Ma
  Oct 14, 20261s
  More actions for workflow run 168

• Qwen Image 2.1 with torch.compile triggers a recompile on each run with varying image size

  skipped:
  New Model Request Reply#167:Issue#15027openedbygeekuillaume
  Oct 14, 20262s
  More actions for workflow run 167

• SDXL invisible watermark: apply_watermark encodes float input, and about a third of outputs are not detected by Stability's detect.py

  skipped:
  New Model Request Reply#166:Issue#15025openedbyCyberNativeAI
  Oct 14, 20261s
  More actions for workflow run 166

• Modular pipelines: warn when a denoiser_input_fields entry is not passed to the transformer

  skipped:
  New Model Request Reply#165:Issue#15022openedbyyiyixuxu
  Oct 13, 20261s
  More actions for workflow run 165

• add_noise() before set_timesteps() uses the wrong noise level (DPMSolverMultistep, DPMSolverSinglestep, UniPC, DEIS; also SanaPipeline)

  skipped:
  New Model Request Reply#164:Issue#15021openedbyFilipLangiewicz
  Oct 13, 20267s
  More actions for workflow run 164

• [LoRA] Text-encoder LoRA loading assumes CLIP module names and fails for T5 / SmolLM3 / ChatGLM text encoders

  skipped:
  New Model Request Reply#163:Issue#15020openedbySergasgr
  Oct 12, 202611s
  More actions for workflow run 163

• offload_state_dict=True leaves CPU model weights on meta after loading

  skipped:
  New Model Request Reply#162:Issue#15018openedbyShenrui-Ma
  Oct 12, 2026Skipped
  More actions for workflow run 162

• gfx1201 (RDNA4) SageAttention attention backend: aiter package or Hub kernel?

  skipped:
  New Model Request Reply#161:Issue#15014openedbyShan2L
  Oct 12, 20261s
  More actions for workflow run 161

• VaeImageProcessor.denormalize() raises AttributeError for NumPy inputs

  skipped:
  New Model Request Reply#160:Issue#15011openedbySrijanRoy123-github
  Oct 12, 20261s
  More actions for workflow run 160