gpt-oss 120B — B200 vs MI355X
Head-to-head AI inference benchmark comparison of B200 (NVIDIA Blackwell) 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.
Throughput at 104 tok/s/user on gpt-oss 120B: B200 hits 32860 tok/s/chip, MI355X hits 23358. Per-million costs land at $0.01 and $0.02 respectively. B200 is 22% cheaper per token; B200 delivers 41% more tok/s/chip.
B200 / MI355X on gpt-oss 120B at 171 tok/s/user: 20448 / 13182 tok/s/chip, $0.02 / $0.03 per million tokens. B200 is 35% cheaper per token; B200 delivers 55% more tok/s/chip.
Toward the upper edge of the 38–305 tok/s/user interactivity band, at 238 tok/s/user on gpt-oss 120B: B200 runs 13725 tok/s/chip at $0.04/M tokens, MI355X runs 3911 at $0.11/M. B200 is 204% cheaper per token; B200 delivers 251% more tok/s/chip. (Numbers reflect the default 8k/1k · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B200:32859.9MI355X:23358.2 | B200:20448.4MI355X:13181.8 | B200:13724.8MI355X:3911.2 |
| Cost ($/M tok) | B200:$0.015MI355X:$0.018 | B200:$0.024MI355X:$0.032 | B200:$0.035MI355X:$0.107 |
| tok/s/MW | B200:19216342MI355X:11176178 | B200:11958138MI355X:6307067 | B200:8026222MI355X:1871375 |
| Concurrency | B200:~33MI355X:~26 | B200:~30MI355X:~9 | B200:~6MI355X:~16 |
Inference Performance
Agentic inference metrics from the AgentX scenario and fixed-sequence inference metrics across models, hardware configurations, and serving parameters.
Total Tokens per $1 USD (Owning - Hyperscaler) vs. Interactivity
gpt-oss 120B • FP4 • 8K / 1K • Source: SemiAnalysis InferenceX™
TCO $/chip/hr: H100: 1.17H200: 1.22B200: 1.73B300: 2.26GB200: 1.86GB300: 2.31MI300X: 0.95MI325X: 1.1MI355X: 1.5RTX6000PRO: 0.68
Source: SemiAnalysis Market July 2026 Pricing Surveys & AI Cloud TCO Model
Note: Disaggregated inference configurations (e.g., MoRI SGLang, Dynamo TRTLLM) calculate cost per million tokens per decode chip or per prefill chip, rather than per total chip count. This makes direct token cost comparison with aggregated configs not an apples-to-apples comparison.
Note: Disaggregated inference configurations (e.g., MoRI SGLang, Dynamo TRTLLM) calculate input throughput per decode chip or per prefill chip, rather than per total chip count. This makes direct input throughput comparison with aggregated configs not an apples-to-apples comparison.
Note: Disaggregated inference configurations (e.g., MoRI SGLang, Dynamo TRTLLM) calculate output throughput per decode chip or per prefill chip, rather than per total chip count. This makes direct output throughput comparison with aggregated configs not an apples-to-apples comparison.
Note: Disaggregated inference configurations (e.g., MoRI SGLang, Dynamo TRTLLM) calculate power per decode chip or per prefill chip, rather than per total chip count. This makes direct power comparison with aggregated configs not an apples-to-apples comparison.
Note: Disaggregated inference configurations (e.g., MoRI SGLang, Dynamo TRTLLM) calculate Joules per decode chip or per prefill chip, rather than per total chip count. This makes direct Joules per token comparison with aggregated configs not an apples-to-apples comparison.
No data available
Please change the model, sequence, precision, date range or chip selection.
Shift+Scroll to zoom • Drag to pan • Double-click to reset • Click a point to pin tooltip