-
Notifications
You must be signed in to change notification settings - Fork 7.3k
All issues
Issue creation is restricted in this repository
Issues
is:issue state:open
is:issue state:open
Search results
- Status: Open.#14784 In huggingface/diffusers;
Deprecating
Siglip2ImageProcessorFastbugSomething isn't workingSomething isn't workingneeds-code-exampleWaiting for relevant code example to be providedWaiting for relevant code example to be providedStatus: Open.#14780 In huggingface/diffusers;MLU backend utilities fail despite MLU device detection
bugSomething isn't workingSomething isn't workingStatus: Open.#14779 In huggingface/diffusers;LTX2 import fails with Transformers 5.5.0 due to annotation-only Gemma4Unified dependency
bugSomething isn't workingSomething isn't workingStatus: Open.#14773 In huggingface/diffusers;[Bug / Optimization] Make AutoencoderKLWan stateless to fix torch.compile compatibility and enable 1.45x speedup
bugSomething isn't workingSomething isn't workingStatus: Open.#14770 In huggingface/diffusers;save_pretrained(safe_serialization=False) leaves the old safetensors checkpoint behind, and from_pretrained loads it instead of the new weights
bugSomething isn't workingSomething isn't workingStatus: Open.#14769 In huggingface/diffusers;Cosmos2VideoToWorldPipelineapplies FPS RoPE modulation that the released Cosmos-Predict2 checkpoints disable natively (default fps=16 → temporal RoPE ×1.5)bugSomething isn't workingSomething isn't workingStatus: Open.#14768 In huggingface/diffusers;- Status: Open.#14761 In huggingface/diffusers;
Pipeline properties that read attributes __call__ never sets (SD3 skip_guidance_layers, LEdits++ XL guidance_scale)
bugSomething isn't workingSomething isn't workingStatus: Open.#14759 In huggingface/diffusers;- Status: Open.#14749 In huggingface/diffusers;
MiniMax-H3 VAE decoder loads in fp32 but CUDA decode runs it in fp16
bugSomething isn't workingSomething isn't workingStatus: Open.- Status: Open.#14744 In huggingface/diffusers;
