FastH3 ships official quantized weights down to 8GB VRAM, Ref2VA reference support coming in V3
pilkyton · reddit · 2026-10-07
FastVideo's open-weight FastH3 V2 project has a status update. Released three weeks ago, V2 is widely regarded as near-parity with the full MiniMax H3 on quality and motion consistency—an impressive feat for a small lab backed by NVIDIA and others—and the community got it running on single consumer GPUs on launch day.
Today the team shipped official quantized weights for various consumer GPUs, plus a new Trim model for cards with as little as 8GB VRAM. Limitations: Ref2VA (reference images/videos/audio to video) and first/last-frame support weren't distilled, and difficult motion, fine detail, and some audio remain below base H3. The good news: the team confirmed "Omni Ref" (Ref2VA) is the next focus, so reference-aware distilled weights are coming in FastH3 V3.
More from Multimodal
- AI video contest winner reimagines Louis XIV's court colonizing Mars — Loo_Atreides · 2026-10-07
- AI comic magazine TischLog #48 explores reversed morality, built with Midjourney V8.2 — tisch_eins · 2026-10-07
- MEND: RL for flow models via proximal velocity matching beats Flow-GRPO in 100 vs ~4k updates — UTEXAS · 2026-10-07
- JLD: perceptual distance from a frozen encoder's Jacobian, fitted in 35s from 100 images, beats LPIPS and DISTS — Shreshth Saini · 2026-10-07
- Qwen-Image Multiple-Angles LoRA released on Hugging Face — Illustrious_Row_9971 · 2026-10-07
- Dev ships Vulkan-only local generative art app with no telemetry or cloud — ogimaru · 2026-10-07