Muse Spark Open 816B — H100 vs MI355X Performance per Dollar
Cost per million tokens of H100 (NVIDIA Hopper) versus MI355X (AMD CDNA 4) on Muse Spark Open 816B. Large-hyperscaler-volume ownership TCO normalized by total tokens — performance per dollar across LLM workloads. Pick the more cost-efficient SKU at every target interactivity level. Use the chart controls below to switch sequences, precisions, and metrics — 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 →
At 64 tok/s/user on Muse Spark Open 816B, H100 costs $0.25 per million tokens; MI355X costs $0.04. MI355X is 450% more cost-efficient at this operating point.
MI355X edges H100 at 98 tok/s/user on Muse Spark Open 816B — $0.07 per million tokens versus $0.25, a 264% cost-per-token gap.
Push Muse Spark Open 816B to 132 tok/s/user and H100 lands at $0.25 per million tokens against MI355X's $0.12 — MI355X pulls ahead by 104%. (Numbers reflect the default agentic-traces · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
Chip pricing (owning hyperscaler): H100 $1.17/chip/hr · MI355X $1.50/chip/hr. Source: SemiAnalysis Market July 2026 Pricing Surveys & AI Cloud TCO Model.
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Dollar per Million Tokens | H100:$0.247MI355X:$0.045 | H100:$0.247MI355X:$0.068 | H100:$0.247MI355X:$0.121 |
| Concurrency | H100:~1MI355X:~16 | H100:~1MI355X:~11 | H100:~1MI355X:~2 |
Inference Performance
Agentic inference metrics from the AgentX scenario and fixed-sequence inference metrics across models, hardware configurations, and serving parameters.
No data available
No measurements to plot for this selection. Review the benchmark controls above or adjust quick filters.