XiaomiAPI Host BenchmarksEstimated
MiMo V2.5 Pro — API Providers & Speed Matrix
Modelled throughput, TTFT latency and prompt-cache pricing estimates across cloud hosting endpoints, derived from official list prices and public benchmarks.
Fastest provider
813 t/s
Lowest latency (TTFT)
90 ms P50
Best price
$0.19 / $0.43
Blended Pricing Workload Calculator
Switch between standard conversational ratios, agentic caching loops, and heavy RAG workloads.
| Provider & Host | Hardware Stack | Output Speed (TPS) | Latency TTFT (P50) | Input Price ($/1M) | Output Price ($/1M) | Cache Read ($/1M) | Blended Price ($/1M) | SLA |
|---|---|---|---|---|---|---|---|---|
XiaomiOfficial | Cloud TPU / Nvidia H100 | 65 t/s | 2.74 s (3.84 s p90) | $0.43 | $0.87 | $0.043 | $0.54 | 99.99% |
| Groq LPU v2 (SRAM) | 273 t/s | 90 ms (120 ms p90) | $0.37 | $0.74 | $0.034 | $0.46 | 99.95% | |
| CS-3 Wafer-Scale Engine | 813 t/s | 120 ms (160 ms p90) | $0.39 | $0.78 | — | $0.49 | 99.90% | |
| Nvidia H100 SXM5 | 98 t/s | 2.19 s (3.01 s p90) | $0.28 | $0.61 | $0.026 | $0.36 | 99.95% | |
| Nvidia H100 / L40S | 85 t/s | 2.47 s (3.43 s p90) | $0.19 | $0.43 | $0.022 | $0.25 | 99.90% | |
| Nvidia B200 / H100 | 124 t/s | 1.92 s (2.60 s p90) | $0.26 | $0.57 | $0.026 | $0.34 | 99.99% |
Figures are modelled estimates based on published list prices and public throughput benchmarks — not measured on live endpoints. Verify current pricing and terms with each host before committing.