Kimi K2.5/K2.6/K2.7-Code 1T · GPU comparison

Kimi K2.5/K2.6/K2.7-Code 1T — GB300 NVL72 vs H200

Head-to-head AI inference benchmark comparison of GB300 NVL72 (NVIDIA Blackwell) and H200 (NVIDIA Hopper) 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.

GB300 NVL72: 5218 tok/s/GPU, $0.14 per million tokens at 50 tok/s/user on Kimi K2.5/K2.6/K2.7-Code 1T. H200 is unmeasured here.

At 85 tok/s/user on Kimi K2.5/K2.6/K2.7-Code 1T, GB300 NVL72 delivers 471 tok/s/GPU at $1.52 per million tokens; H200 hasn't been benchmarked at this target.

GB300 NVL72 hits 234 tok/s/GPU for $3.14 per million tokens at 119 tok/s/user on Kimi K2.5/K2.6/K2.7-Code 1T. No H200 data at this operating point. (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)
GB300 NVL72:5218.4H200:
GB300 NVL72:471.3H200:
GB300 NVL72:233.8H200:
Cost ($/M tok)
GB300 NVL72:$0.138H200:
GB300 NVL72:$1.524H200:
GB300 NVL72:$3.138H200:
tok/s/MW
GB300 NVL72:2484948H200:
GB300 NVL72:224425H200:
GB300 NVL72:111347H200:
Concurrency
GB300 NVL72:~1324H200:
GB300 NVL72:~95H200:
GB300 NVL72:~34H200:

Inference Performance

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