2.4K views · 44 likes · 15 reposts (at collection)
vLLM adds day-zero support for MiniMax M3.1 Flash
- Who
- vLLM
- What
- adds Day-0 support for MiniMax M3.1 Flash in the vLLM serving stack
What happened
vLLM announces day-zero support for MiniMax M3.1 Flash from @MiniMax_AI. The post describes multi-step automation, real-codebase engineering, and visual reasoning, with native NVFP4 KV, NVFP4 MoE weights, and DSpark speculative decoding in the serving stack.
Links in the post: 2Why it matters
The announcement connects day-one model availability with specific serving features, giving deployment teams concrete context for evaluating MiniMax M3.1 Flash support in vLLM.
Read next
See the full storyMiniMax M3.1 Flash Launches on Ollama Cloud, Gains vLLM Support and Inferact Production ServingThe story so far
Ollama added MiniMax M3.1 Flash to its cloud service, vLLM added support in its serving stack, and vLLM says Inferact is serving the model in production at scale with guaranteed SLAs. MiniMax said it is partnering with Inferact on production inference for MiniMax models, with engineering improvements contributed back to vLLM. Ollama says its deployment is hosted in the US with zero data retention and can be used through Ollama Cloud, Claude Code and Codex. It describes the model as supporting coding, agentic workflows, image understanding and a 1M-token context window. In collaboration with MiniMax, Ollama announced a three-day free offer for the model on Ollama Cloud on October 20. vLLM says its support includes native NVFP4 KV, NVFP4 MoE weights and DSpark speculative decoding.
Keep up with AI, one email a day
Model releases, price changes and compute developments, with original sources. Free, confirmed by email, and easy to unsubscribe. Read a past issue · 2026-10-23