Qwen 3.5 397B-A17B · B200 · Precision comparison
B200: FP4 vs FP8 precision comparison
How FP4 and FP8 precision affect Qwen 3.5 397B-A17B 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.
Throughput at 97 tok/s/user on Qwen 3.5 397B-A17B (B200): FP4 hits 11,230 tok/s/chip, FP8 hits 6,990. Costs per million tokens land at $0.043 and $0.069 respectively. FP4 delivers 61% more tok/s/chip on the same chip, which makes it 38% cheaper per token. The cost-throughput tradeoff from lower precision is only part of the picture — see the evaluation page for accuracy data.
Around the middle of the 10–358 tok/s/user interactivity band, at 184 tok/s/user on Qwen 3.5 397B-A17B (B200): FP4 runs 6,498 tok/s/chip at $0.074 per million tokens, FP8 runs 4,143 at $0.12. FP4 delivers 57% more tok/s/chip on the same chip, which makes it 36% cheaper per token. Precision changes affect both inference speed and model quality — consult the evaluation tab for accuracy benchmarks.
At 271 tok/s/user on Qwen 3.5 397B-A17B (B200), FP4 delivers 4,545 tok/s/chip at $0.11 per million tokens; FP8 delivers 2,663 tok/s/chip at $0.18. FP4 delivers 71% more tok/s/chip on the same chip, which makes it 41% cheaper per token. Lower-precision quantization trades model accuracy for throughput — check the evaluation page for quality impact. (Numbers reflect the default 8k/1k selection for this URL — table and chart below update if you change sequence or model in the controls. Each side uses the best available serving configuration for that precision, which may include speculative decoding such as MTP where recipes exist — the same convention as the other comparison pages.)
Interpolated from real benchmark data. Edit target interactivity values below to compare at different operating points.
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | FP411,230FP86,990 | FP46,498FP84,143 | FP44,545FP82,663 |
| Cost ($/M tok) | FP4$0.043FP8$0.069 | FP4$0.074FP8$0.12 | FP4$0.11FP8$0.18 |
| tok/s/MW | FP46,567,057FP84,087,629 | FP43,800,290FP82,422,884 | FP42,658,016FP81,557,301 |
| Concurrency | FP4~28FP8~36 | FP4~9FP8~11 | FP4~4FP8~4 |
