MiniMaxLaunchAI ModelsAugust 3, 2026

MiniMax releases H3 open-weight video model

MiniMax H3 generates up to 15s of 2K video with native stereo sound and supports multi-image to video, image to video, and video editing tasks. The model is available via ComfyUI, LumaLabs AI Agents, Leonardo.ai, and Picsart, with commercial deployment licensing available for the US, EU, UK, and South Korea.

How this story unfolded

13 days · 4 reports · 53 community posts · 57 of 64 shown

Jul 29
- Weights aren't the only thing that should be open-source. Both kernel repos behind our recent work...
Jul 31
- MiniMax H3 now available on Vercel AI Gatewayvercel.com
Aug 1
- MiniMax Releases MiniMax H3: An Omni-Modal Video Model That Generates 15-Second 2K Clips With Native Stereo Audiomarktechpost.com
Aug 2
Aug 3
Aug 4
Aug 5
- Open weights = access → possibility. Open communities. Better performance. Deeper customization....
Aug 6
Aug 8
- [Minimax H3] If you’re on a Blackwell GPU (RTX 50-series), upgrade to PyTorch 2.13 and CUDA 13 ASAP (the performance gains are massive)
Aug 9
- Minimax H3 is so good at complex camera movments
Aug 10
Aug 11

MiniMax by email

Get an email when MiniMax has news

No news that day, no email.

MiniMax releases H3 open-weight video model

How this story unfolded

MiniMax by email

More stories today