GLM 5/5.1 · Chip comparison
GLM 5/5.1 — H200 vs MI355X
Head-to-head AI inference benchmark comparison of H200 (NVIDIA Hopper) and MI355X (AMD CDNA 4) 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.
Near the low end of the 18–84 tok/s/user interactivity band, at 34 tok/s/user on GLM 5/5.1: H200 runs 740 tok/s/chip at $0.46 per million tokens, MI355X runs 1,276 at $0.33. MI355X is 29% cheaper per token; MI355X delivers 72% more tok/s/chip.
Setting 51 tok/s/user as the target on GLM 5/5.1, H200 produces 563 tok/s/chip ($0.60 per million tokens) and MI355X produces 957 ($0.44). MI355X is 28% cheaper per token; MI355X delivers 70% more tok/s/chip.
At 68 tok/s/user interactivity on GLM 5/5.1, H200 delivers 434 tok/s/chip at $0.78 per million tokens; MI355X delivers 747 tok/s/chip at $0.56. MI355X is 29% cheaper per token; MI355X delivers 72% more tok/s/chip at this point. (Numbers reflect the default 8k/1k · fp8 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
Interpolated from real benchmark data. Edit target interactivity values below to compare at different operating points.
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | H200740MI355X1,276 | H200563MI355X957 | H200434MI355X747 |
| Cost ($/M tok) | H200$0.46MI355X$0.33 | H200$0.60MI355X$0.44 | H200$0.78MI355X$0.56 |
| tok/s/MW | H200540,170MI355X610,522 | H200411,175MI355X457,666 | H200316,738MI355X357,251 |
| Concurrency | H200~21MI355X~18 | H200~10MI355X~9 | H200~6MI355X~5 |