Diffusers

This release introduces several major image and video pipelines (Flux2, Hunyuan 1.5, Sana-Video), high-performance attention backends via the 'kernels' library, and the TaylorSeer caching method for significant speed improvements.

This patch release focuses on bug fixes for model initialization, device placement for Wan and LTX models, and ensuring compatibility with PyTorch 2.3.1.

This release introduces improvements to the Qwen-Image Edit model within the diffusers library.

This release introduces major new pipelines (Wan 2.2, Flux-Kontext, Qwen-Image), significant performance optimizations via regional compilation and GGUF CUDA kernels, and an experimental modular pipeline system.

This release introduces several major video and image generation pipelines including Wan VACE, Cosmos Predict2, and Chroma, while significantly improving torch.compile performance and memory optimization techniques.

This release provides a bug fix for the Wan pipeline to correctly handle ftfy imports.

This release introduces several major video and image generation pipelines including Wan2.1, LTX Video 0.9.5, and Lumina2, alongside significant memory optimization features like Layerwise Casting and Group Offloading.

This patch release fixes regressions in Flux single-file and LoRA loading (including 4-bit support) and resolves several bugs in HunyuanVideo related to batch processing and LoRA compatibility.