GLM 5/5.1 — B300 vs MI355X
Head-to-head AI inference benchmark comparison of B300 (NVIDIA Blackwell) and MI355X (AMD CDNA 4) on GLM 5/5.1. 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.
B300 posts 1409 tok/s/chip for $0.45 per million tokens at 27 tok/s/user on GLM 5/5.1; MI355X posts 1619 tok/s/chip for $0.26. MI355X is 73% cheaper per token; MI355X delivers 15% more tok/s/chip.
Throughput at 48 tok/s/user on GLM 5/5.1: B300 hits 1122 tok/s/chip, MI355X hits 988. Per-million costs land at $0.56 and $0.42 respectively. MI355X is 33% cheaper per token; B300 delivers 14% more tok/s/chip.
B300 / MI355X on GLM 5/5.1 at 69 tok/s/user: 874 / 735 tok/s/chip, $0.72 / $0.57 per million tokens. MI355X is 27% cheaper per token; B300 delivers 19% more tok/s/chip. (Numbers reflect the default 8k/1k · fp8 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B300:1409.4MI355X:1619.2 | B300:1122.2MI355X:988.4 | B300:874.1MI355X:734.5 |
| Cost ($/M tok) | B300:$0.445MI355X:$0.257 | B300:$0.559MI355X:$0.422 | B300:$0.718MI355X:$0.567 |
| tok/s/MW | B300:741793MI355X:774737 | B300:590617MI355X:472912 | B300:460069MI355X:351454 |
| Concurrency | B300:~48MI355X:~27 | B300:~21MI355X:~10 | B300:~12MI355X:~5 |
Inference Performance
Agentic inference metrics from the AgentX scenario and fixed-sequence inference metrics across models, hardware configurations, and serving parameters.
Total Tokens per $1 USD (Owning - Hyperscaler) vs. Interactivity
GLM5/5.1 744B • FP4 • 8K / 1K • Source: SemiAnalysis InferenceX™
TCO $/chip/hr: H100: 1.17H200: 1.22B200: 1.73B300: 2.26GB200: 1.86GB300: 2.31MI300X: 0.95MI325X: 1.1MI355X: 1.5RTX6000PRO: 0.68
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
Please change the model, sequence, precision, date range or chip selection.
Shift+Scroll to zoom • Drag to pan • Double-click to reset • Click a point to pin tooltip