GLM 5.3 744B — GB200 NVL72 vs MI355X Performance per Dollar
Cost per million tokens of GB200 NVL72 (NVIDIA Blackwell) versus MI355X (AMD CDNA 4) on GLM 5.3 744B. 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 137 tok/s/user on GLM 5.3 744B, GB200 NVL72 costs $0.04 per million tokens; MI355X costs $0.04. GB200 NVL72 is 10% more cost-efficient at this operating point.
GB200 NVL72 edges MI355X at 182 tok/s/user on GLM 5.3 744B — $0.05 per million tokens versus $0.08, a 58% cost-per-token gap.
Push GLM 5.3 744B to 228 tok/s/user and GB200 NVL72 lands at $0.09 per million tokens against MI355X's $0.11 — GB200 NVL72 pulls ahead by 19%. (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): GB200 NVL72 $1.86/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 | GB200 NVL72:$0.040MI355X:$0.044 | GB200 NVL72:$0.053MI355X:$0.084 | GB200 NVL72:$0.091MI355X:$0.109 |
| Concurrency | GB200 NVL72:~69MI355X:~5 | GB200 NVL72:~46MI355X:~2 | GB200 NVL72:~6MI355X:~4 |
Inference Performance
Agentic inference metrics from the AgentX scenario and fixed-sequence inference metrics across models, hardware configurations, and serving parameters.
GLM5.2/GLM5.3 744B 8K / 1K Cost per Million Total Tokens vs. Interactivity
GLM5.2/GLM5.3 744B 8K / 1K Cost per Million Total Tokens vs. Interactivity- Cost Tier:
- Source:
- SemiAnalysis InferenceX™
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 data available
No measurements to plot for this selection. Review the benchmark controls above or adjust quick filters.
Shift+Scroll to zoom • Drag to pan • Double-click to reset • Click a point to pin tooltip