Head-to-head AI inference benchmark comparison of B200 (NVIDIA Blackwell) and MI355X (AMD CDNA 4) on Llama 3.3 70B. 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 37 tok/s/user on Llama 3.3 70B: B200 hits 6594 tok/s/chip, MI355X hits 4326. Per-million costs land at $0.07 and $0.10 respectively. B200 is 32% cheaper per token; B200 delivers 52% more tok/s/chip.
B200 / MI355X on Llama 3.3 70B at 61 tok/s/user: 4854 / 2613 tok/s/chip, $0.10 / $0.16 per million tokens. B200 is 61% cheaper per token; B200 delivers 86% more tok/s/chip.
Toward the upper edge of the 14–108 tok/s/user interactivity band, at 85 tok/s/user on Llama 3.3 70B: B200 runs 3538 tok/s/chip at $0.14/M tokens, MI355X runs 1060 at $0.39/M. B200 is 189% cheaper per token; B200 delivers 234% more tok/s/chip. (Numbers reflect the default 8k/1k · fp8 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B200:6593.5MI355X:4326.2 | B200:4854.3MI355X:2612.5 | B200:3537.9MI355X:1059.7 |
| Cost ($/M tok) | B200:$0.073MI355X:$0.096 | B200:$0.099MI355X:$0.159 | B200:$0.136MI355X:$0.393 |
| tok/s/MW | B200:3855848MI355X:2069965 | B200:2838777MI355X:1250014 | B200:2068956MI355X:507051 |
| Concurrency | B200:~44MI355X:~18 | B200:~32MI355X:~22 | B200:~21MI355X:~6 |
Agentic inference metrics from the AgentX scenario and fixed-sequence inference metrics across models, hardware configurations, and serving parameters.
TCO $/chip/hr: VR200: 3.61H100: 1.17H200: 1.22B200: 1.73B300: 2.26GB200: 1.86GB300: 2.31MI300X: 0.95MI325X: 1.1MI355X: 1.5RTX6000PRO: 0.68Jalapeño (Teacup): 1.47
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.
Matching measurements exist, but their chip series are hidden.
Shift+Scroll to zoom • Drag to pan • Double-click to reset • Click a point to pin tooltip
Llama 3.3 70B Instruct 8K / 1K Total Tokens per $1 TCO vs. Interactivity