vLLM Announces Day 0 Support for Kimi K3 Across NVIDIA Architectures

vllm_project · x · 2026-07-30

The vLLM project has announced first-class support for the Kimi K3 model, optimized for NVIDIA hardware.

The deployment serves K3 across Grace Blackwell, Blackwell, Hopper, and NVL72 systems. It also integrates with Dynamo to maintain speed and efficiency for such a large model at production scale. This support was available on Day 0, the moment the model weights went public.

Related event: vLLM and AMD Announce Day-0 Inference Support for Kimi K3(7 posts)→

Original post →

More from Infra

Infra channel →