FastVideo releases FastH3 V1: 4-step sparse distilled LoRA
Solitary_Thinker · reddit · 2026-08-29
FastVideo team released FastH3 V1 for MINIMAX h3, featuring 4-step sparse distilled checkpoints and LoRAs to improve speed and quality.
Details:
- VSA Version: Requires VSA kernel;
- Dense Version: Easy-to-test dense attention LoRA;
- Training: 1,000+ B200 hours used;
- Features: Native support for variable resolution, aspect ratio, and duration.
Plans: Fully open source code with training recipes coming soon. Optimizations for RTX, DGX Sparks, and Apple MLX are planned, along with integration of Parallel Decoding Distillation (PDD).
More from Multimodal
- VideoGen-Agent: An 8B Multimodal Agent Turns Video Generation Into a Tool-Use Loop — jiqizhixin · 2026-10-16
- Drama-3 weights land on the Hugging Face Hub with an official demo — Gradio · 2026-10-16
- MiniCPM-V 4.5 offer open-source local image understanding to replace paid vision APIs — heyshrutimishra · 2026-10-16
- Surreal AI-generated landscape hides "another landscape inside the measurements" — piotrbinkowski · 2026-10-16
- techhalla: Building 3D scenes in Blender is the key to AI filmmaking — techhalla · 2026-10-16
- Fan animation made in Blender with Claude Code (Opus 5.5) — Icy-Appointment3583 · 2026-10-16