gpt-oss 120B — H200 vs MI355X
Head-to-head AI inference benchmark comparison of H200 (NVIDIA Hopper) and MI355X (AMD CDNA 4) on gpt-oss 120B. 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 71–270 tok/s/user interactivity band, at 120 tok/s/user on gpt-oss 120B: H200 runs 2720 tok/s/GPU at $0.14/M tokens, MI355X runs 8848 at $0.05/M. MI355X is 212% cheaper per token; MI355X delivers 225% more tok/s/GPU.
Setting 170 tok/s/user as the target on gpt-oss 120B, H200 produces 1489 tok/s/GPU ($0.26 per million tokens) and MI355X produces 4137 ($0.10). MI355X is 166% cheaper per token; MI355X delivers 178% more tok/s/GPU.
At 220 tok/s/user interactivity on gpt-oss 120B, H200 delivers 759 tok/s/GPU at $0.52 per million tokens; MI355X delivers 2238 tok/s/GPU at $0.18. MI355X is 185% cheaper per token; MI355X delivers 195% more tok/s/GPU at this point. (Numbers reflect the default 1k/1k · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/gpu) | H200:2720.0MI355X:8848.3 | H200:1489.5MI355X:4136.6 | H200:759.4MI355X:2238.0 |
| Cost ($/M tok) | H200:$0.144MI355X:$0.046 | H200:$0.262MI355X:$0.099 | H200:$0.522MI355X:$0.183 |
| tok/s/MW | H200:1572262MI355X:3338976 | H200:860978MI355X:1560967 | H200:438949MI355X:844545 |
| Concurrency | H200:~64MI355X:~38 | H200:~39MI355X:~12 | H200:~7MI355X:~5 |
Inference Performance
Inference performance metrics across different models, hardware configurations, and serving parameters.