Head-to-head AI inference benchmark comparison of B300 (NVIDIA Blackwell) and MI325X (AMD CDNA 3) on GLM 5.3 744B. 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 →
B300 posts 11227 tok/s/chip for $0.06 per million tokens at 60 tok/s/user on GLM 5.3 744B; MI325X posts 1208 tok/s/chip for $0.25. B300 is 352% cheaper per token; B300 delivers 829% more tok/s/chip.
Throughput at 119 tok/s/user on GLM 5.3 744B: B300 hits 8377 tok/s/chip, MI325X hits 1206. Per-million costs land at $0.07 and $0.25 respectively. B300 is 238% cheaper per token; B300 delivers 595% more tok/s/chip.
B300 / MI325X on GLM 5.3 744B at 178 tok/s/user: 4601 / 1206 tok/s/chip, $0.14 / $0.25 per million tokens. B300 is 86% cheaper per token; B300 delivers 282% more tok/s/chip. (Numbers reflect the default agentic-traces · fp8 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B300:11227.3MI325X:1208.3 | B300:8377.1MI325X:1205.8 | B300:4601.0MI325X:1205.8 |
| Cost ($/M tok) | B300:$0.056MI325X:$0.253 | B300:$0.075MI325X:$0.253 | B300:$0.136MI325X:$0.253 |
| tok/s/MW | B300:5909109MI325X:714970 | B300:4409010MI325X:713473 | B300:2421587MI325X:713473 |
| Concurrency | B300:~16MI325X:~1 | B300:~13MI325X:~1 | B300:~7MI325X:~1 |
Agentic inference metrics from the AgentX scenario and fixed-sequence inference metrics across models, hardware configurations, and serving parameters.
No measurements to plot for this selection. Review the benchmark controls above or adjust quick filters.
