MiniMax H3 lands in vLLM with an OpenAI-compatible video endpoint
- Source
- x.com
- Date
Open weights are step one. Making them easy to serve is step two. H3 now has Day 0 support in @vllm_project's vLLM-Omni, complete with an OpenAI-compatible video endpoint. Huge thanks to the vLLM team for bringing H3 into the open inference stack. ⚡ #MiniMaxH3 #OpenWeights https://t.co/01dKD8PngX
Context
MiniMax H3's integration with vLLM-Omni gives developers a serving interface for the released model weights. The project's announcement describes an OpenAI-compatible /v1/videos endpoint that can return a result directly or let an application check a generation job's progress. The output is an MP4 containing video and synchronized stereo sound, so the application receives both together. This supplies the deployment step beyond H3's model announcement.
Checking sign-in…
Loading comments…