GLM 5/5.1 · GPU comparison

GLM 5/5.1 — H200 vs MI325X

Head-to-head AI inference benchmark comparison of H200 (NVIDIA Hopper) 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.

H200 posts 396 tok/s/GPU for $0.99 per million tokens at 24 tok/s/user on GLM 5/5.1; MI325X posts 148 tok/s/GPU for $2.46. H200 is 149% cheaper per token; H200 delivers 168% more tok/s/GPU.

Throughput at 28 tok/s/user on GLM 5/5.1: H200 hits 396 tok/s/GPU, MI325X hits 75. Per-million costs land at $0.99 and $4.82 respectively. H200 is 387% cheaper per token; H200 delivers 431% more tok/s/GPU.

H200 / MI325X on GLM 5/5.1 at 31 tok/s/user: 396 / 54 tok/s/GPU, $0.99 / $6.55 per million tokens. H200 is 562% cheaper per token; H200 delivers 627% more tok/s/GPU. (Numbers reflect the default 1k/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 view →

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/gpu)
H200:395.7MI325X:147.5
H200:395.7MI325X:74.5
H200:395.7MI325X:54.5
Cost ($/M tok)
H200:$0.990MI325X:$2.465
H200:$0.990MI325X:$4.823
H200:$0.990MI325X:$6.553
tok/s/MW
H200:228711MI325X:67680
H200:228711MI325X:34179
H200:228711MI325X:24978
Concurrency
H200:~64MI325X:~25
H200:~64MI325X:~14
H200:~64MI325X:~8

Inference Performance

Inference performance metrics across different models, hardware configurations, and serving parameters.