MI355X FP8: M3 EAGLE vs Off Speculative Decoding
Speculative decoding comparison of M3 EAGLE versus Off on MI355X FP8 (AMD CDNA 4) running MiniMax M3 428B. 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.
Near the low end of the 8–148 tok/s/user interactivity band, at 43 tok/s/user on MiniMax M3 428B (MI355X FP8): M3 EAGLE runs 6274 tok/s/chip at $0.07/M tokens, Off runs 4374 at $0.10/M. M3 EAGLE is 43% cheaper per token; M3 EAGLE delivers 43% more tok/s/chip. Gains from speculative decoding vary by workload; short-output prompts tend to benefit less.
At 78 tok/s/user on MiniMax M3 428B (MI355X FP8), M3 EAGLE delivers 4409 tok/s/chip at $0.09 per million tokens; Off delivers 2114 tok/s/chip at $0.20. M3 EAGLE is 109% cheaper per token; M3 EAGLE delivers 109% more tok/s/chip. Speculative decoding accepts draft tokens to reduce per-token latency — gains vary by workload and prompt distribution.
M3 EAGLE posts 3359 tok/s/chip for $0.12 per million tokens at 114 tok/s/user on MiniMax M3 428B (MI355X FP8); Off posts 1125 tok/s/chip for $0.37. M3 EAGLE is 199% cheaper per token; M3 EAGLE delivers 199% more tok/s/chip. Draft-token acceptance rates determine whether speculative decoding helps or hurts at a given concurrency level. (Numbers reflect this URL's pinned 8k/1k · fp8 workload — changing sequence or model updates both the table and chart; the table stays pinned to this page's precision, so precision toggles in the controls affect the chart only.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | M3 EAGLE:6274.3Off:4374.5 | M3 EAGLE:4409.3Off:2114.3 | M3 EAGLE:3359.3Off:1125.0 |
| Cost ($/M tok) | M3 EAGLE:$0.066Off:$0.095 | M3 EAGLE:$0.094Off:$0.197 | M3 EAGLE:$0.124Off:$0.370 |
| tok/s/MW | M3 EAGLE:3002063Off:2093040 | M3 EAGLE:2109720Off:1011611 | M3 EAGLE:1607312Off:538256 |
| Concurrency | M3 EAGLE:~78Off:~95 | M3 EAGLE:~30Off:~12 | M3 EAGLE:~16Off:~5 |
Inference Performance
Agentic inference metrics from the AgentX scenario and fixed-sequence inference metrics across models, hardware configurations, and serving parameters.