MiniMax H3 Max tops video leaderboards via fal's post-training
ArtificialAnlys · x · 2026-08-27
MiniMax H3 Max, a post-trained version of MiniMax H3 developed by fal, debuts at #1 in Image to Video and #3 in Text to Video on the Artificial Analysis Video Leaderboards with Audio. Tuned for stronger prompt adherence and better aesthetics, it generates 5-15 second clips with native audio at up to 768p. Priced at $0.04 per second, fal also intends to release the weights, which would make it the highest ranked open weights model on the boards.
More from Models
- Users Report GPT-5.6 Quality Drop Ahead of Potential Astra Release — Realistic_Stomach848 · 2026-08-27
- Claude tightens undocumented generation restrictions, impacting design outputs — max_paperclips · 2026-08-27
- Peter Yang tests ChatGPT, Claude, Grok and Gemini across 10 use cases — nickbaumann_ · 2026-08-27
- MiniMax H3 at Ray Summit: Showcasing 33B Open-Weight Audio-Video Model — MiniMax_AI · 2026-08-27
- GLM-5.3 Flash Review: GPT-5.6 Level Performance at Ultra-Low Cost — zainhas · 2026-08-27
- Qwen 3.8 and GLM 5.3 Flash open models released, available for Dell on-prem deployment — _akhaliq · 2026-08-27