MiniMax H3 inpainting goes live on Hugging Face, open-model video editing at its best
linoy_tsaban · x · 2026-08-22
MiniMax H3's inpainting feature is now live on a Hugging Face Space. Usage: just prompt what you want to inpaint (with auto mask generation) or click the subject to replace; optionally add reference images/video for motion control; keep original audio or upload/generate new. The author says video editing with open models has never looked this good. A community port to diffusers modular blocks with custom blocks and inference examples is also available, tested with a turbo LoRA.
More from coding & agent
- Vercel Goes Full Backend: Python GA, Containers, and Daemons — cramforce · 2026-10-16
- WorldMemArena Accepted at NeurIPS: Memory Storage ≠ Memory Use — xwang_lk · 2026-10-16
- Vercel Ship Demos Container Deploys for Go, Ruby, Java and More — edwin · 2026-10-16
- TRL adds day-0 support for fine-tuning Muse Spark Open's 816B MoE with LoRA on 64 H100s — QGallouedec · 2026-10-16
- Databricks engineers land impactful PRs on day 2 with AI dev tools, vs week 3 before — Yuchenj_UW · 2026-10-16
- Perplexity open-sources SPACE, the sandbox platform powering its Computer agents — AravSrinivas · 2026-10-16