# CoreWeave offers Kimi K3 deployment on dedicated inference

_Published Monday, August 31, 2026 at 8:18 PM EDT · AI, Infrastructure · Latest · Tier 2 — Notable_

![CoreWeave offers Kimi K3 deployment on dedicated inference — Primary](https://jf.x.com/images/post/2094404129917452402.png)

CoreWeave said customers can now deploy Kimi K3 through its Dedicated Inference service by bringing their own model weights. The company said the service deploys the weights on Nvidia GPUs and handles autoscaling and routing. CoreWeave described Kimi K3 as a 2.8-trillion-parameter open-source model with a one-million-token context window.

## Sources

- [CoreWeave (@CoreWeave)](https://x.com/CoreWeave/status/2094404129917452402)

---
Canonical: https://techandbusiness.org/newswire/qFwlKrYyqbC58OuA35x5f4
Published: 2026-09-01T00:18:05.986Z
Story chronology: 2026-08-31T12:37:11.000Z
Retrieved: 2026-10-16T09:19:36.402Z
Publisher: Tech & Business (techandbusiness.org)