Kimi K3 runs on AMD MI355X at 952 tok/s per node, claim says
AccBalanced · x · 2026-08-04
A quoted post says the team got Kimi K3 running on AMD MI355X at 952 tok/s/node and 118 tok/s single-stream.
The claim also says the setup delivers 3.8× the aggregate throughput per node and 1.3× the single-stream decode of B200, while beating B300 on performance per dollar at 48 vs. 33 tok/s/$. More details were promised in replies.
Related event: AMD MI355X Runs Kimi K3: Nearly 4x B200 Throughput, 71% Cost Drop(6 posts)→
More from Infra
- AMD Faces AI Earnings Test as Open Models Threaten Closed Labs' Margins — AccBalanced · 2026-08-04
- AMD MI355X runs Kimi K3 on single node, 3.8x throughput of B200 setup, better cost-performance than B300 — 机器之心 · 2026-08-04
- MiniMax H3 Text-to-Video Successfully Runs on a Single DGX Spark — Scobleizer · 2026-08-04
- Spectrum Acceleration for MiniMax H3 in ComfyUI: Up to 34% Lower Inference Time — marres · 2026-08-04
- Counterpoint Research: DRAM and NAND Memory Prices Expected to Peak in 2027 — SumitGup · 2026-08-04
- Atomic-scale logic circuits built from silicon dangling bonds showcased in new paper — teortaxesTex · 2026-08-04