Inferact says it co-led the optimization effort with Red Hat AI, Nvidia and Huawei.
Inferact reported a 2.2–2.8× speedup for Kimi K3 on vLLM in its September 18 announcement. It says the work spans scheduling, KDA state handling and custom mixture-of-experts (MoE) kernels, and shared a technical deep dive into the optimizations.
—
4 Sources, first seen ago
—
Not ranked yet
—
Not ranked yet
@vllm_projectKimi K3 serving in vLLM now delivers 2.2–2.8x throughput on our B300 benchmark vs v0.27.1. We break down the work across scheduling, KDA state handling, and MoE kernels, with benchmarks and commands to reproduce the results. Thanks to the vLLM community for pushing Kimi K3 performance forward! Read the deep dive: https://vllm.ai/blog/2026-09-13-kimi-k3-performance-optimization
@inferactKimi K3 on @vllm_project is now 2.2–2.8× faster 🚀 Inferact is proud to have co-led this optimization effort with @RedHat_AI, @NVIDIAAI, and @Huawei, spanning scheduling, KDA state handling, and custom MoE kernels. Read the technical deep dive: https://vllm.ai/blog/2026-09-13-kimi-k3-performance-optimization
@woosuk_kRT @inferact: Kimi K3 on @vllm_project is now 2.2–2.8× faster 🚀 Inferact is proud to have co-led this optimization effort with @RedHat_AI,…@vllm_projectKimi K3 serving in vLLM now delivers 2.2–2.8x throughput on our B300 benchmark vs v0.27.1. We break down the work across scheduling, KDA state handling, and MoE kernels, with benchmarks and commands to reproduce the results. Thanks to the vLLM community for pushing Kimi K3 performance forward! Read the deep dive: https://vllm.ai/blog/2026-09-13-kimi-k3-performance-optimization@inferactKimi K3 on @vllm_project is now 2.2–2.8× faster 🚀 Inferact is proud to have co-led this optimization effort with @RedHat_AI, @NVIDIAAI, and @Huawei, spanning scheduling, KDA state handling, and custom MoE kernels. Read the technical deep dive: https://vllm.ai/blog/2026-09-13-kimi-k3-performance-optimization@woosuk_kRT @inferact: Kimi K3 on @vllm_project is now 2.2–2.8× faster 🚀 Inferact is proud to have co-led this optimization effort with @RedHat_AI,…@vllm_projectKimi K3 serving in vLLM now delivers 2.2–2.8x throughput on our B300 benchmark vs v0.27.1. We break down the work across scheduling, KDA state handling, and MoE kernels, with benchmarks and commands to reproduce the results. Thanks to the vLLM community for pushing Kimi K3 performance forward! Read the deep dive: https://vllm.ai/blog/2026-09-13-kimi-k3-performance-optimization@inferactKimi K3 on @vllm_project is now 2.2–2.8× faster 🚀 Inferact is proud to have co-led this optimization effort with @RedHat_AI, @NVIDIAAI, and @Huawei, spanning scheduling, KDA state handling, and custom MoE kernels. Read the technical deep dive: https://vllm.ai/blog/2026-09-13-kimi-k3-performance-optimization@woosuk_kRT @inferact: Kimi K3 on @vllm_project is now 2.2–2.8× faster 🚀 Inferact is proud to have co-led this optimization effort with @RedHat_AI,…@vllm_projectKimi K3 serving in vLLM now delivers 2.2–2.8x throughput on our B300 benchmark vs v0.27.1. We break down the work across scheduling, KDA state handling, and MoE kernels, with benchmarks and commands to reproduce the results. Thanks to the vLLM community for pushing Kimi K3 performance forward! Read the deep dive: https://vllm.ai/blog/2026-09-13-kimi-k3-performance-optimization@inferactKimi K3 on @vllm_project is now 2.2–2.8× faster 🚀 Inferact is proud to have co-led this optimization effort with @RedHat_AI, @NVIDIAAI, and @Huawei, spanning scheduling, KDA state handling, and custom MoE kernels. Read the technical deep dive: https://vllm.ai/blog/2026-09-13-kimi-k3-performance-optimization@woosuk_kRT @inferact: Kimi K3 on @vllm_project is now 2.2–2.8× faster 🚀 Inferact is proud to have co-led this optimization effort with @RedHat_AI,…
