Head-to-head AI inference benchmark comparison of MI355X (AMD CDNA 4) and Vera Rubin NVL72 (NVIDIA Vera Rubin) on GLM 5.3 744B. Latency, throughput, and cost across LLM workloads. Use the chart controls below to switch sequences, precisions, and metrics β same interactions as the main inference chart.
AgentX replays real coding-agent sessions rather than fixed-length prompts, so context grows turn over turn and most of each request is served from cache instead of being recomputed. That turns the comparison into a systems question: KV transfer between nodes, prefix-aware routing, and cache capacity all move the curve alongside raw chip throughput. Fixed-sequence workloads stay the clean baseline for kernel and silicon performance, so the two scenarios answer different questions about the same hardware. Learn more about AgentX β
Near the low end of the 112β271 tok/s/user interactivity band, at 151 tok/s/user on GLM 5.3 744B: MI355X runs 7675 tok/s/chip at $0.05/M tokens, Vera Rubin NVL72 runs 88564 at $0.01/M. Vera Rubin NVL72 is 379% cheaper per token; Vera Rubin NVL72 delivers 1054% more tok/s/chip.
Setting 191 tok/s/user as the target on GLM 5.3 744B, MI355X produces 4488 tok/s/chip ($0.09 per million tokens) and Vera Rubin NVL72 produces 62615 ($0.02). Vera Rubin NVL72 is 480% cheaper per token; Vera Rubin NVL72 delivers 1295% more tok/s/chip.
At 231 tok/s/user interactivity on GLM 5.3 744B, MI355X delivers 4029 tok/s/chip at $0.10 per million tokens; Vera Rubin NVL72 delivers 41542 tok/s/chip at $0.02. Vera Rubin NVL72 is 328% cheaper per token; Vera Rubin NVL72 delivers 931% more tok/s/chip at this point. (Numbers reflect the default agentic-traces Β· fp4 selection for this URL β table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | MI355X:7675.4Vera Rubin NVL72:88563.6 | MI355X:4487.9Vera Rubin NVL72:62615.5 | MI355X:4029.4Vera Rubin NVL72:41541.8 |
| Cost ($/M tok) | MI355X:$0.054Vera Rubin NVL72:$0.011 | MI355X:$0.093Vera Rubin NVL72:$0.016 | MI355X:$0.103Vera Rubin NVL72:$0.024 |
| tok/s/MW | MI355X:3672453Vera Rubin NVL72:26837459 | MI355X:2147330Vera Rubin NVL72:18974391 | MI355X:1927936Vera Rubin NVL72:12588415 |
| Concurrency | MI355X:~4Vera Rubin NVL72:~318 | MI355X:~2Vera Rubin NVL72:~199 | MI355X:~4Vera Rubin NVL72:~108 |
Agentic inference metrics from the AgentX scenario and fixed-sequence inference metrics across models, hardware configurations, and serving parameters.
TCO $/chip/hr:
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 measurements to plot for this selection. Review the benchmark controls above or adjust quick filters.
Shift+Scroll to zoom β’ Drag to pan β’ Double-click to reset β’ Click a point to pin tooltip
GLM5.2/GLM5.3 744B 8K / 1K Total Tokens per $1 TCO vs. Median Interactivity