gpt-oss 120B · Chip comparison

gpt-oss 120B — GB200 NVL72 vs MI300X

Head-to-head AI inference benchmark comparison of GB200 NVL72 (NVIDIA Blackwell) and MI300X (AMD CDNA 3) 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.

At 113 tok/s/user interactivity on gpt-oss 120B, GB200 NVL72 delivers 42,816 tok/s/chip at $0.012 per million tokens; MI300X delivers 4,178 tok/s/chip at $0.063. GB200 NVL72 is 81% cheaper per token; GB200 NVL72 delivers 10× the tok/s/chip at this point.

GB200 NVL72 posts 33,220 tok/s/chip for $0.016 per million tokens at 157 tok/s/user on gpt-oss 120B; MI300X posts 2,916 tok/s/chip for $0.090. GB200 NVL72 is 83% cheaper per token; GB200 NVL72 delivers 11× the tok/s/chip.

Throughput at 202 tok/s/user on gpt-oss 120B: GB200 NVL72 hits 26,214 tok/s/chip, MI300X hits 996. Costs per million tokens land at $0.020 and $0.26 respectively. GB200 NVL72 is 93% cheaper per token; GB200 NVL72 delivers 26× the tok/s/chip. (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.)

View performance per dollar →

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)
GB200 NVL7242,816MI300X4,178
GB200 NVL7233,220MI300X2,916
GB200 NVL7226,214MI300X996
Cost ($/M tok)
GB200 NVL72$0.012MI300X$0.063
GB200 NVL72$0.016MI300X$0.090
GB200 NVL72$0.020MI300X$0.26
tok/s/MW
GB200 NVL7222,896,080MI300X3,006,041
GB200 NVL7217,764,960MI300X2,097,905
GB200 NVL7214,018,263MI300X716,772
Concurrency
GB200 NVL72~375MI300X~9
GB200 NVL72~317MI300X~4
GB200 NVL72~90MI300X~5