DeepSeekv4 Pro 0813 1.6T — B300 vs MI355X
Head-to-head AI inference benchmark comparison of B300 (NVIDIA Blackwell) and MI355X (AMD CDNA 4) on DeepSeekv4 Pro 0813 1.6T. 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.
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 →
Setting 33 tok/s/user as the target on DeepSeekv4 Pro 0813 1.6T, B300 produces 47657 tok/s/chip ($0.01 per million tokens) and MI355X produces 20684 ($0.02). B300 is 53% cheaper per token; B300 delivers 130% more tok/s/chip.
At 61 tok/s/user interactivity on DeepSeekv4 Pro 0813 1.6T, B300 delivers 18041 tok/s/chip at $0.03 per million tokens; MI355X delivers 7400 tok/s/chip at $0.06. B300 is 62% cheaper per token; B300 delivers 144% more tok/s/chip at this point.
B300 posts 12303 tok/s/chip for $0.05 per million tokens at 89 tok/s/user on DeepSeekv4 Pro 0813 1.6T; MI355X posts 3859 tok/s/chip for $0.11. B300 is 112% cheaper per token; B300 delivers 219% more tok/s/chip. (Numbers reflect the default agentic-traces · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B300:47657.2MI355X:20684.1 | B300:18041.4MI355X:7400.3 | B300:12302.8MI355X:3858.7 |
| Cost ($/M tok) | B300:$0.013MI355X:$0.020 | B300:$0.035MI355X:$0.056 | B300:$0.051MI355X:$0.108 |
| tok/s/MW | B300:25082720MI355X:9896677 | B300:9495468MI355X:3540810 | B300:6475151MI355X:1846282 |
| Concurrency | B300:~116MI355X:~149 | B300:~15MI355X:~28 | B300:~9MI355X:~6 |
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
DeepSeek V4 Pro 1.6T • 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