Diffusers 0.40.0 ships stable Modular Diffusers, tensor-parallel, LTX2.5 & MiniMax H3
RisingSayak · x · 2026-08-21
Hugging Face released Diffusers 0.40.0. Highlights:
- Modular Diffusers graduates out of experimental into stable support.
- New pipelines: LTX2.5, MiniMax H3, Music 3, Wan Animate 2. MiniMax-H3 jointly denoises video and soundtrack with a single transformer (no separate vocoder, no post-hoc audio pass), using a Qwen3VL conditioner that reads the 50th decoder layer's hidden state.
- First tensor-parallel support for a select few models.
- New quantization backends: SDNQ, Nunchaku-Lite.
- A massive test redesign — leaner and sometimes faster.
More from Multimodal
- Veo and Kling Lead: Comparison of 15 AI Video Tools — Pale_Intern5543 · 2026-08-21
- User Plays Pictionary Against Google Gemma 4 — tristanbob · 2026-08-21
- 4DAnyone: Create 4D Humans from a Single Casual Video — AntResearch · 2026-08-21
- User creates Mega Man map style LoRA for Krea 2 pixel art generation — Neggy5 · 2026-08-21
- Seeking stylized models for 3D renders convertible to STL printing — MAD_MakeADev · 2026-08-21
- Seedance 2.5 vs Kling 3.0 Omni: A Comparison of Expressiveness and Character Interaction — SimplyAnnisa · 2026-08-21