Kimi K2.5/K2.6 1T · GPU comparison

Kimi K2.5/K2.6 1T — B300 vs MI355X

Head-to-head AI inference benchmark comparison of B300 (NVIDIA Blackwell) and MI355X (AMD CDNA 4) on Kimi K2.5/K2.6 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.

At 61 tok/s/user interactivity on Kimi K2.5/K2.6 1T, B300 delivers 871 tok/s/GPU at $0.74 per million tokens; MI355X delivers 589 tok/s/GPU at $0.70. MI355X is 6% cheaper per token; B300 delivers 48% more tok/s/GPU at this point.

B300 posts 616 tok/s/GPU for $1.06 per million tokens at 79 tok/s/user on Kimi K2.5/K2.6 1T; MI355X posts 489 tok/s/GPU for $0.84. MI355X is 26% cheaper per token; B300 delivers 26% more tok/s/GPU.

Throughput at 96 tok/s/user on Kimi K2.5/K2.6 1T: B300 hits 465 tok/s/GPU, MI355X hits 379. Per-million costs land at $1.37 and $1.09 respectively. MI355X is 26% cheaper per token; B300 delivers 23% more tok/s/GPU. (Numbers reflect the default 1k/1k · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)

View performance-per-dollar view →

Interpolated from real benchmark data. Edit target interactivity values below to compare at different operating points.
Metric
Interactivity (tok/s/user)
Interactivity (tok/s/user)
Interactivity (tok/s/user)
Throughput (tok/s/gpu)
B300:870.6MI355X:589.4
B300:615.8MI355X:488.6
B300:465.4MI355X:378.7
Cost ($/M tok)
B300:$0.743MI355X:$0.699
B300:$1.055MI355X:$0.839
B300:$1.371MI355X:$1.086
tok/s/MW
B300:401191MI355X:222400
B300:283773MI355X:184368
B300:214492MI355X:142918
Concurrency
B300:~29MI355X:~20
B300:~16MI355X:~12
B300:~10MI355X:~8

Inference Performance

Inference performance metrics across different models, hardware configurations, and serving parameters.