H200 FP8: MTP vs Off Speculative Decoding
Speculative decoding comparison of MTP versus Off on H200 FP8 (NVIDIA Hopper) running DeepSeek R1. Throughput, cost, and interactivity differences across LLM workloads. Use the chart controls below to switch sequences and metrics — same interactions as the main inference chart.
MTP acceptance-rate implementations differ across inference engines. Points from different engines are not directly comparable on the same curve — throughput and cost at matched interactivity may reflect engine-level differences rather than pure speculative decoding gains. Interpret cross-engine comparisons with caution.
MTP posts 1832 tok/s/chip for $0.18 per million tokens at 39 tok/s/user on DeepSeek R1 (H200 FP8); Off posts 947 tok/s/chip for $0.36. MTP is 93% cheaper per token; MTP delivers 93% more tok/s/chip. Draft-token acceptance rates determine whether speculative decoding helps or hurts at a given concurrency level.
Throughput at 62 tok/s/user on DeepSeek R1 (H200 FP8): MTP hits 1121 tok/s/chip, Off hits 711. Per-million costs land at $0.30 and $0.48 respectively. MTP is 58% cheaper per token; MTP delivers 58% more tok/s/chip. Speculative decoding trades extra compute on draft tokens for fewer decoding steps — the payoff depends on sequence length and batch size.
Toward the upper edge of the 18–106 tok/s/user interactivity band, at 84 tok/s/user on DeepSeek R1 (H200 FP8): MTP runs 637 tok/s/chip at $0.53/M tokens, Off runs 501 at $0.68/M. MTP is 27% cheaper per token; MTP delivers 27% more tok/s/chip. Gains from speculative decoding vary by workload; short-output prompts tend to benefit less. (Numbers reflect this URL's pinned 8k/1k · fp8 workload — changing sequence or model updates both the table and chart; the table stays pinned to this page's precision, so precision toggles in the controls affect the chart only.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | MTP:1832.1Off:947.4 | MTP:1121.4Off:711.3 | MTP:637.4Off:501.5 |
| Cost ($/M tok) | MTP:$0.185Off:$0.358 | MTP:$0.302Off:$0.476 | MTP:$0.532Off:$0.676 |
| tok/s/MW | MTP:1337271Off:691499 | MTP:818567Off:519194 | MTP:465254Off:366056 |
| Concurrency | MTP:~128Off:~22 | MTP:~160Off:~11 | MTP:~7Off:~5 |
Inference Performance
Agentic inference metrics from the AgentX scenario and fixed-sequence inference metrics across models, hardware configurations, and serving parameters.