MiniMax M3 428B · Chip comparison
MiniMax M3 428B — MI300X vs MI325X
Head-to-head AI inference benchmark comparison of MI300X (AMD CDNA 3) and MI325X (AMD CDNA 3) on MiniMax M3 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 →
MI300X / MI325X on MiniMax M3 428B at 28 tok/s/user: 5,025 / 940 tok/s/chip, $0.053 / $0.33 per million tokens. MI300X is 84% cheaper per token; MI300X delivers 5.3× the tok/s/chip.
Around the middle of the 28–30 tok/s/user interactivity band, at 29 tok/s/user on MiniMax M3 428B: MI300X runs 5,025 tok/s/chip at $0.053 per million tokens, MI325X runs 912 at $0.34. MI300X is 84% cheaper per token; MI300X delivers 5.5× the tok/s/chip.
Setting 30 tok/s/user as the target on MiniMax M3 428B, MI300X produces 5,025 tok/s/chip ($0.053 per million tokens) and MI325X produces 882 ($0.35). MI300X is 85% cheaper per token; MI300X delivers 5.7× the tok/s/chip. (Numbers reflect the default agentic-traces · fp8 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
Interpolated from real benchmark data. Edit target interactivity values below to compare at different operating points.
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | MI300X5,025MI325X940 | MI300X5,025MI325X912 | MI300X5,025MI325X882 |
| Cost ($/M tok) | MI300X$0.053MI325X$0.33 | MI300X$0.053MI325X$0.34 | MI300X$0.053MI325X$0.35 |
| tok/s/MW | MI300X3,615,012MI325X556,192 | MI300X3,615,012MI325X539,380 | MI300X3,615,012MI325X521,728 |
| Concurrency | MI300X~10MI325X~2 | MI300X~10MI325X~2 | MI300X~10MI325X~1 |