Kimi K2.5/K2.6/K2.7-Code 1T · B200 · Precision comparison
B200: FP4 vs INT4 precision comparison
How FP4 and INT4 precision affect Kimi K2.5/K2.6/K2.7-Code 1T inference on B200 (NVIDIA Blackwell). Throughput, latency, and cost across LLM workloads. Use the chart controls below to switch sequences and metrics — same interactions as the main inference chart.
FP4 posts 8,817 tok/s/chip for $0.055 per million tokens at 49 tok/s/user on Kimi K2.5/K2.6/K2.7-Code 1T (B200); INT4 posts 1,459 tok/s/chip for $0.33. FP4 delivers 6.0× the tok/s/chip on the same chip, which makes it 83% cheaper per token. Quantization-level accuracy differences are tracked on the evaluation tab.
Throughput at 68 tok/s/user on Kimi K2.5/K2.6/K2.7-Code 1T (B200): FP4 hits 3,630 tok/s/chip, INT4 hits 1,016. Costs per million tokens land at $0.13 and $0.47 respectively. FP4 delivers 3.6× the tok/s/chip on the same chip, which makes it 72% cheaper per token. The cost-throughput tradeoff from lower precision is only part of the picture — see the evaluation page for accuracy data.
Toward the upper edge of the 30–105 tok/s/user interactivity band, at 87 tok/s/user on Kimi K2.5/K2.6/K2.7-Code 1T (B200): FP4 runs 1,968 tok/s/chip at $0.24 per million tokens, INT4 runs 655 at $0.73. FP4 delivers 3.0× the tok/s/chip on the same chip, which makes it 67% cheaper per token. Precision changes affect both inference speed and model quality — consult the evaluation tab for accuracy benchmarks. (Numbers reflect the default 8k/1k selection for this URL — table and chart below update if you change sequence or model in the controls. Each side uses the best available serving configuration for that precision, which may include speculative decoding such as MTP where recipes exist — the same convention as the other comparison pages.)
Interpolated from real benchmark data. Edit target interactivity values below to compare at different operating points.
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | FP48,817INT41,459 | FP43,630INT41,016 | FP41,968INT4655 |
| Cost ($/M tok) | FP4$0.055INT4$0.33 | FP4$0.13INT4$0.47 | FP4$0.24INT4$0.73 |
| tok/s/MW | FP45,155,980INT4852,937 | FP42,122,686INT4594,390 | FP41,151,001INT4382,858 |
| Concurrency | FP4~309INT4~26 | FP4~171INT4~13 | FP4~123INT4~7 |
