AMD MI355X vLLM Beats Nvidia B200 on Kimi K2.5 Inference
marksaroufim · x · 2026-08-02
Recent benchmarks from SemiAnalysis reveal that AMD's MI355X has outperformed Nvidia's B200 in vLLM inference running Kimi K2.5 (sharing the same architecture as xAI Cursor Composer 2.5).
This performance leap is largely attributed to upstream AMD kernels optimized by the @GPUMODE community, highlighting the growing impact of open-source collaboration in maximizing hardware inference efficiency.
More from Infra
- DeepSeek's New Release Significantly Boosts the Value of Nvidia DGX Spark — firstadopter · 2026-08-02
- Developer Showcases Running Hermes Model Locally on Dell Mini PC — burhop · 2026-08-02
- State of AI Compute Index: Anthropic and OpenAI Shift Heavily to Non-Nvidia Chips — nathanbenaich · 2026-08-02
- Maryland County Passes 18-Month Moratorium on Data Center Construction — LadyGagas913 · 2026-08-02
- Power Shortage Becomes the New Bottleneck for the AI Race Beyond Chips — ingliguori · 2026-08-02
- Together AI's Monthly Token Volume Hits 400 Trillion, Marking 10,000x Growth — togethercompute · 2026-08-02