Head-to-head AI inference benchmark comparison of B300 (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 →
Near the low end of the 27–204 tok/s/user interactivity band, at 71 tok/s/user on DeepSeek V4.1 Flash 552B: B300 runs 114372 tok/s/chip at $0.01/M tokens, MI300X runs 8784 at $0.03/M. B300 is 447% cheaper per token; B300 delivers 1202% more tok/s/chip.
Setting 116 tok/s/user as the target on DeepSeek V4.1 Flash 552B, B300 produces 58081 tok/s/chip ($0.01 per million tokens) and MI300X produces 5131 ($0.05). B300 is 376% cheaper per token; B300 delivers 1032% more tok/s/chip.
At 160 tok/s/user interactivity on DeepSeek V4.1 Flash 552B, B300 delivers 53743 tok/s/chip at $0.01 per million tokens; MI300X delivers 3787 tok/s/chip at $0.07. B300 is 496% cheaper per token; B300 delivers 1319% more tok/s/chip at this point. (Numbers reflect the default agentic-traces · fp4 selection for this URL — table and chart below update if you change sequence, precision, or model in the controls.)
| Metric | Interactivity (tok/s/user) | Interactivity (tok/s/user) | Interactivity (tok/s/user) |
|---|---|---|---|
| Throughput (tok/s/chip) | B300:114372.4MI300X:8783.7 | B300:58080.7MI300X:5131.1 | B300:53743.0MI300X:3787.4 |
| Cost ($/M tok) | B300:$0.005MI300X:$0.030 | B300:$0.011MI300X:$0.051 | B300:$0.012MI300X:$0.070 |
| tok/s/MW | B300:60195985MI300X:6319223 | B300:30568792MI300X:3691417 | B300:28285774MI300X:2724781 |
| Concurrency | B300:~45MI300X:~7 | B300:~44MI300X:~3 | B300:~16MI300X:~1 |
Agentic inference metrics from the AgentX scenario and fixed-sequence inference metrics across models, hardware configurations, and serving parameters.
No measurements to plot for this selection. Review the benchmark controls above or adjust quick filters.
