GLM 5/5.1 — GB200 NVL72 vs MI325X
Head-to-head AI inference benchmark comparison of GB200 NVL72 (NVIDIA Blackwell) and MI325X (AMD CDNA 3) on GLM 5/5.1. 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.
GB200 NVL72: 9135 tok/s/chip, $0.06 per million tokens at 67 tok/s/user on GLM 5/5.1. MI325X is unmeasured here.
At 104 tok/s/user on GLM 5/5.1, GB200 NVL72 delivers 3619 tok/s/chip at $0.14 per million tokens; MI325X hasn't been benchmarked at this target.
GB200 NVL72 hits 630 tok/s/chip for $0.82 per million tokens at 141 tok/s/user on GLM 5/5.1. No MI325X data at this operating point. (Numbers reflect the default 8k/1k · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | GB200 NVL72:9135.4MI325X:— | GB200 NVL72:3618.9MI325X:— | GB200 NVL72:629.8MI325X:— |
| Cost ($/M tok) | GB200 NVL72:$0.057MI325X:— | GB200 NVL72:$0.143MI325X:— | GB200 NVL72:$0.820MI325X:— |
| tok/s/MW | GB200 NVL72:4885238MI325X:— | GB200 NVL72:1935265MI325X:— | GB200 NVL72:336816MI325X:— |
| Concurrency | GB200 NVL72:~698MI325X:— | GB200 NVL72:~231MI325X:— | GB200 NVL72:~62MI325X:— |
Inference Performance
Agentic inference metrics from the AgentX scenario and fixed-sequence inference metrics across models, hardware configurations, and serving parameters.