Kimi K3 runs on vLLM + AMD from Day 0, supporting 2.8T params on Instinct
vllm_project · x · 2026-07-30
The vLLM project announced that the Kimi K3 model natively supports vLLM and AMD's ROCm ecosystem from day zero.
This allows teams to deploy the full 2.8T parameter model out-of-the-box on AMD Instinct hardware. Broader performance tuning for Instinct is currently in progress.
Related event: Kimi K3 Gets Day-0 Support for vLLM and AMD ROCm(2 posts)→
More from Infra
- Cognition Lab Talk: RL and Inference Optimization Are Converging — AAAzzam · 2026-07-30
- Vector Institute Demystifies MoE: Slashes Logit Memory from 23.3GB to 0.3GB — VectorInst · 2026-07-30
- Deploying LTX Video Models on Cloud GPUs: Pitfalls and an Automated Installer — Humble_Cut6799 · 2026-07-30
- NVIDIA Expected to Raise GeForce RTX GPU Prices Again by Up to 30% — ANR2ME · 2026-07-30
- Edge AI Strategy: Trading 10% Accuracy for 15x Energy Efficiency — JLeonsarmiento · 2026-07-30
- Cloud Revenue Growth Showdown: Google Cloud Hits 63% in Q1 2026, Beating Azure and AWS — Beth_Kindig · 2026-07-30