Head-to-head AI inference benchmark comparison of GB300 NVL72 (NVIDIA Blackwell) and MI355X (AMD CDNA 4) on MiniMax M2.5/M2.7. 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.
GB300 NVL72 posts 10654 tok/s/chip for $0.06 per million tokens at 43 tok/s/user on MiniMax M2.5/M2.7; MI355X posts 4697 tok/s/chip for $0.09. GB300 NVL72 is 47% cheaper per token; GB300 NVL72 delivers 127% more tok/s/chip.
Throughput at 62 tok/s/user on MiniMax M2.5/M2.7: GB300 NVL72 hits 7309 tok/s/chip, MI355X hits 3138. Per-million costs land at $0.09 and $0.13 respectively. GB300 NVL72 is 51% cheaper per token; GB300 NVL72 delivers 133% more tok/s/chip.
GB300 NVL72 / MI355X on MiniMax M2.5/M2.7 at 81 tok/s/user: 3698 / 2173 tok/s/chip, $0.17 / $0.19 per million tokens. GB300 NVL72 is 11% cheaper per token; GB300 NVL72 delivers 70% more tok/s/chip. (Numbers reflect the default 8k/1k · fp8 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | GB300 NVL72:10653.9MI355X:4697.0 | GB300 NVL72:7308.7MI355X:3138.1 | GB300 NVL72:3698.0MI355X:2172.7 |
| Cost ($/M tok) | GB300 NVL72:$0.060MI355X:$0.089 | GB300 NVL72:$0.088MI355X:$0.133 | GB300 NVL72:$0.174MI355X:$0.192 |
| tok/s/MW | GB300 NVL72:5025439MI355X:2247388 | GB300 NVL72:3447521MI355X:1501464 | GB300 NVL72:1744352MI355X:1039561 |
| Concurrency | GB300 NVL72:~200MI355X:~43 | GB300 NVL72:~197MI355X:~11 | GB300 NVL72:~51MI355X:~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
1 The ATOM engine is promising, however it has yet to serve production tokens. It is still in its infant stage.
