Head-to-head AI inference benchmark comparison of MI355X (AMD CDNA 4) and Vera Rubin NVL72 (NVIDIA Vera Rubin) on Qwen 3.8 Flash Next 176B-A6B. 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
Throughput at 86 tok/s/user on Qwen 3.8 Flash Next 176B-A6B: MI355X hits 58839 tok/s/chip, Vera Rubin NVL72 hits 252442. Per-million costs land at $0.01 and $0.00 respectively. Vera Rubin NVL72 is 78% cheaper per token; Vera Rubin NVL72 delivers 329% more tok/s/chip.
MI355X / Vera Rubin NVL72 on Qwen 3.8 Flash Next 176B-A6B at 106 tok/s/user: 36602 / 234872 tok/s/chip, $0.01 / $0.00 per million tokens. Vera Rubin NVL72 is 167% cheaper per token; Vera Rubin NVL72 delivers 542% more tok/s/chip.
Toward the upper edge of the 67β145 tok/s/user interactivity band, at 125 tok/s/user on Qwen 3.8 Flash Next 176B-A6B: MI355X runs 19807 tok/s/chip at $0.02/M tokens, Vera Rubin NVL72 runs 219422 at $0.00/M. Vera Rubin NVL72 is 360% cheaper per token; Vera Rubin NVL72 delivers 1008% more tok/s/chip. (Numbers reflect the default agentic-traces Β· fp4 selection for this URL β table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | MI355X:58838.9Vera Rubin NVL72:252442.2 | MI355X:36602.3Vera Rubin NVL72:234872.5 | MI355X:19807.3Vera Rubin NVL72:219421.7 |
| Cost ($/M tok) | MI355X:$0.007Vera Rubin NVL72:$0.004 | MI355X:$0.011Vera Rubin NVL72:$0.004 | MI355X:$0.021Vera Rubin NVL72:$0.005 |
| tok/s/MW | MI355X:28152565Vera Rubin NVL72:76497626 | MI355X:17513060Vera Rubin NVL72:71173476 | MI355X:9477154Vera Rubin NVL72:66491434 |
| Concurrency | MI355X:~13Vera Rubin NVL72:~207 | MI355X:~8Vera Rubin NVL72:~234 | MI355X:~5Vera Rubin NVL72:~239 |
Agentic inference metrics from the AgentX scenario and fixed-sequence inference metrics across models, hardware configurations, and serving parameters.
