diffusers · Issues· 1428 open
Open on GitHubLocally synced open issues (discussions stay on GitHub)
- #14803
53 messages in src/diffusers are missing the f-string prefix (safety-checker ValueError in 46 pipelines)
Updated Sep 17, 2026 - #14802
Revisit modular diffusers weight-duplication workaround now that huggingface_hub has a shared blobs cache
Updated Sep 17, 2026 - #14700
Add VideoDeltaNet-H3 Model
Updated Sep 17, 2026 - #14794
TangentialClassifierFreeGuidance.is_conditional reads an attribute that does not exist
bugpipelinesneeds-env-infoUpdated Sep 16, 2026 - #10076
Do we have any script covert from hf format to orginal format?
good first issuecontributions-welcomeconversion scriptUpdated Sep 16, 2026 - #14779
MLU backend utilities fail despite MLU device detection
bugmodelsneeds-env-infoUpdated Sep 16, 2026 - #14784
Opt-in weight integrity verification at load time (pinned SHA-256 hashes / manifest)
pipelinesfeature-requestUpdated Sep 15, 2026 - #14773
LTX2 import fails with Transformers 5.5.0 due to annotation-only Gemma4Unified dependency
bugpipelinesneeds-env-infoUpdated Sep 15, 2026 - #14768
`Cosmos2VideoToWorldPipeline` applies FPS RoPE modulation that the released Cosmos-Predict2 checkpoints disable natively (default fps=16 → temporal RoPE ×1.5)
bugmodelspipelinesneeds-env-infoUpdated Sep 15, 2026 - #14780
Deprecating `Siglip2ImageProcessorFast`
bugneeds-code-examplepipelinesneeds-env-infoUpdated Sep 15, 2026 - #14769
save_pretrained(safe_serialization=False) leaves the old safetensors checkpoint behind, and from_pretrained loads it instead of the new weights
bugpipelinesUpdated Sep 15, 2026 - #14770
[Bug / Optimization] Make AutoencoderKLWan stateless to fix torch.compile compatibility and enable 1.45x speedup
bugmodelsUpdated Sep 14, 2026 - #14746
MiniMax-H3 VAE decoder loads in fp32 but CUDA decode runs it in fp16
bugmodular-pipelinesneeds-env-infoUpdated Sep 14, 2026 - #14761
Add IP-Adapter support to StableDiffusionXLInstructPix2PixPipeline
pipelinesfeature-requestUpdated Sep 13, 2026 - #14473
Thoughts on maintaining our lightweight Docker container to reduce startup time for remote inference?
pipelinesfeature-requestUpdated Sep 13, 2026