MiniMax M3.1 Flash 428B · Chip comparison

MiniMax M3.1 Flash 428B — H100 vs MI355X

Head-to-head AI inference benchmark comparison of H100 (NVIDIA Hopper) and MI355X (AMD CDNA 4) on MiniMax M3.1 Flash 428B. 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.

AgentX replays real coding-agent sessions rather than fixed-length prompts, so context grows turn over turn and most of each request is served from cache instead of being recomputed. That turns the comparison into a systems question: KV transfer between nodes, prefix-aware routing, and cache capacity all move the curve alongside raw chip throughput. Fixed-sequence workloads stay the clean baseline for kernel and silicon performance, so the two scenarios answer different questions about the same hardware. Learn more about AgentX →

H100 posts 5,282 tok/s/chip for $0.062 per million tokens at 117 tok/s/user on MiniMax M3.1 Flash 428B; MI355X posts 111,965 tok/s/chip for $0.004. MI355X is 94% cheaper per token; MI355X delivers 21× the tok/s/chip.

Throughput at 174 tok/s/user on MiniMax M3.1 Flash 428B: H100 hits 4,107 tok/s/chip, MI355X hits 89,805. Costs per million tokens land at $0.079 and $0.005 respectively. MI355X is 94% cheaper per token; MI355X delivers 22× the tok/s/chip.

H100 / MI355X on MiniMax M3.1 Flash 428B at 231 tok/s/user: 2,969 / 69,238 tok/s/chip, $0.11 / $0.006 per million tokens. MI355X is 95% cheaper per token; MI355X delivers 23× the tok/s/chip. (Numbers reflect the default agentic-traces · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)

View performance per dollar →

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)
H1005,282MI355X111,965
H1004,107MI355X89,805
H1002,969MI355X69,238
Cost ($/M tok)
H100$0.062MI355X$0.004
H100$0.079MI355X$0.005
H100$0.11MI355X$0.006
tok/s/MW
H1003,855,453MI355X53,571,673
H1002,998,163MI355X42,968,693
H1002,167,464MI355X33,128,197
Concurrency
H100~8MI355X~63
H100~6MI355X~48
H100~3MI355X~35