FastH3 Omni

FastH3 Omni is a newly tracked term on FastToolHub's radar. Source signal: X+HF+Web. Radar summary: Hao AI Lab (UCSD FastVideo) + Nuva Lab open-sourced FastH3 Omni on 2026-10-20: distilled MiniMax H3 for reference-guided video+audio (text + up to 9 images, 3 videos, 3 audio refs → 5–15s 24fps clips with synced audio) in 8 transformer steps; 5.6–11.9× lower latency on 8×GB200; NVFP4 QAD weights for DGX Spark / RTX 5090; ComfyUI support. HF collection FastVideo/fasth3-omni (Comfy repo 41 likes). Nuva Lab pitches it as open-source 'Unlimited Seedance'. Local AI release post planned Oct 21. X News story.

FastH3 Omni appears in the model / AI-system category: something teams would load, call, or fine-tune rather than a pure UI gadget. Open-source or public-repo signals show up in the excerpt, so inspection and self-hosting may be possible. Do not invent parameters, prices, or benchmarks beyond what the excerpt already states.

Start from the primary link: https://haoailab.com/blogs/fastvideo-omniref-update/. For models: verify context length, license, and hardware needs from official docs; run a small smoke test before production. Numbers and claims in this page only repeat the radar excerpt — double-check upstream before publishing.