MiniMax M2.5/M2.7 · Chip comparison

MiniMax M2.5/M2.7 — H200 vs MI325X

Head-to-head AI inference benchmark comparison of H200 (NVIDIA Hopper) and MI325X (AMD CDNA 3) on MiniMax M2.5/M2.7. 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.

Setting 34 tok/s/user as the target on MiniMax M2.5/M2.7, H200 produces 3,353 tok/s/chip ($0.10 per million tokens) and MI325X produces 2,699 ($0.11). H200 is 11% cheaper per token; H200 delivers 24% more tok/s/chip.

At 56 tok/s/user interactivity on MiniMax M2.5/M2.7, H200 delivers 2,342 tok/s/chip at $0.14 per million tokens; MI325X delivers 1,814 tok/s/chip at $0.17. H200 is 14% cheaper per token; H200 delivers 29% more tok/s/chip at this point.

H200 posts 1,681 tok/s/chip for $0.20 per million tokens at 78 tok/s/user on MiniMax M2.5/M2.7; MI325X posts 1,009 tok/s/chip for $0.30. H200 is 33% cheaper per token; H200 delivers 67% more tok/s/chip. (Numbers reflect the default 8k/1k · fp8 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)

View performance per dollar →

MI325X vs H200: full specs & benchmarks across models →

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)
H2003,353MI325X2,699
H2002,342MI325X1,814
H2001,681MI325X1,009
Cost ($/M tok)
H200$0.10MI325X$0.11
H200$0.14MI325X$0.17
H200$0.20MI325X$0.30
tok/s/MW
H2002,447,382MI325X1,596,901
H2001,709,253MI325X1,073,566
H2001,226,691MI325X597,118
Concurrency
H200~44MI325X~18
H200~19MI325X~8
H200~10MI325X~5