DeepSeek V4.1 Flash 552B — B200 vs MI300X
Head-to-head AI inference benchmark comparison of B200 (NVIDIA Blackwell) and MI300X (AMD CDNA 3) on DeepSeek V4.1 Flash 552B. 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 →
B200 / MI300X on DeepSeek V4.1 Flash 552B at 90 tok/s/user: 357162 / 8684 tok/s/chip, $0.00 / $0.03 per million tokens. B200 is 2159% cheaper per token; B200 delivers 4013% more tok/s/chip.
Around the middle of the 52–205 tok/s/user interactivity band, at 128 tok/s/user on DeepSeek V4.1 Flash 552B: B200 runs 319861 tok/s/chip at $0.00/M tokens, MI300X runs 5923 at $0.04/M. B200 is 2865% cheaper per token; B200 delivers 5300% more tok/s/chip.
Setting 167 tok/s/user as the target on DeepSeek V4.1 Flash 552B, B200 produces 210692 tok/s/chip ($0.00 per million tokens) and MI300X produces 3869 ($0.07). B200 is 2891% cheaper per token; B200 delivers 5346% more tok/s/chip. (Numbers reflect the default agentic-traces · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B200:357161.8MI300X:8683.9 | B200:319860.7MI300X:5923.4 | B200:210691.6MI300X:3868.6 |
| Cost ($/M tok) | B200:$0.001MI300X:$0.030 | B200:$0.002MI300X:$0.045 | B200:$0.002MI300X:$0.068 |
| tok/s/MW | B200:208866557MI300X:6247383 | B200:187053070MI300X:4261410 | B200:123211466MI300X:2783173 |
| Concurrency | B200:~256MI300X:~6 | B200:~220MI300X:~4 | B200:~141MI300X:~1 |
Inference Performance
Agentic inference metrics from the AgentX scenario and fixed-sequence inference metrics across models, hardware configurations, and serving parameters.