Kimi K2.5/K2.6/K2.7-Code 1T — B300 vs MI300X
Head-to-head AI inference benchmark comparison of B300 (NVIDIA Blackwell) and MI300X (AMD CDNA 3) on Kimi K2.5/K2.6/K2.7-Code 1T. Latency, throughput, and cost across LLM workloads. Use the chart controls below to switch sequences, precisions, and metrics — same interactions as the main inference chart.
B300 posts 722 tok/s/GPU for $0.91 per million tokens at 33 tok/s/user on Kimi K2.5/K2.6/K2.7-Code 1T; MI300X posts 94 tok/s/GPU for $3.34. B300 is 267% cheaper per token; B300 delivers 671% more tok/s/GPU.
Throughput at 38 tok/s/user on Kimi K2.5/K2.6/K2.7-Code 1T: B300 hits 621 tok/s/GPU, MI300X hits 76. Per-million costs land at $1.05 and $4.07 respectively. B300 is 288% cheaper per token; B300 delivers 713% more tok/s/GPU.
B300 / MI300X on Kimi K2.5/K2.6/K2.7-Code 1T at 44 tok/s/user: 544 / 60 tok/s/GPU, $1.20 / $5.32 per million tokens. B300 is 344% cheaper per token; B300 delivers 811% more tok/s/GPU. (Numbers reflect the default 1k/1k · int4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/gpu) | B300:722.3MI300X:93.7 | B300:620.6MI300X:76.3 | B300:543.7MI300X:59.7 |
| Cost ($/M tok) | B300:$0.910MI300X:$3.342 | B300:$1.049MI300X:$4.070 | B300:$1.200MI300X:$5.324 |
| tok/s/MW | B300:332852MI300X:52335 | B300:286003MI300X:42648 | B300:250546MI300X:33342 |
| Concurrency | B300:~47MI300X:~12 | B300:~34MI300X:~8 | B300:~23MI300X:~6 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.