Baseten adds day-0 API support for Moonshot’s Kimi K3 on vLLM

AccBalanced · x · 2026-07-28

Baseten says it has day-0 support for Kimi K3 on its Model APIs, and vLLM is collaborating on the rollout.

The linked article walks through how to build a day-0 API for a newly released model: Moonshot shared early-access weights, Baseten worked with Inferact and RadixArk, and the setup is presented as an example of fast inference-stack integration around an open-weight model.

Related event: Baseten and vLLM Deliver Day-0 API for Kimi K3(2 posts)→

Original post →

More from coding & agent

coding & agent channel →