Liquid AIAPI Host BenchmarksEstimated
LFM2.5-8B-A1B — 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
4225 t/s
Lowest latency (TTFT)
90 ms P50
Best price
$0.23 / $1
Blended Pricing Workload Calculator
Switch between standard conversational ratios, agentic caching loops, and heavy RAG workloads.
| Provider & Host | Hardware Stack | Output Speed (TPS) | Latency TTFT (P50) | Input Price ($/1M) | Output Price ($/1M) | Cache Read ($/1M) | Blended Price ($/1M) | SLA |
|---|---|---|---|---|---|---|---|---|
Liquid AIOfficial | Cloud TPU / Nvidia H100 | 338 t/s | 1.48 s (2.07 s p90) | $0.5 | $2 | $0.05 | $0.88 | 99.99% |
| Groq LPU v2 (SRAM) | 1420 t/s | 90 ms (120 ms p90) | $0.42 | $1.7 | $0.04 | $0.74 | 99.95% | |
| CS-3 Wafer-Scale Engine | 4225 t/s | 120 ms (160 ms p90) | $0.45 | $1.8 | — | $0.79 | 99.90% | |
| Nvidia H100 SXM5 | 507 t/s | 1.18 s (1.63 s p90) | $0.33 | $1.4 | $0.03 | $0.6 | 99.95% | |
| Nvidia H100 / L40S | 439 t/s | 1.33 s (1.85 s p90) | $0.23 | $1 | $0.025 | $0.42 | 99.90% | |
| Nvidia B200 / H100 | 642 t/s | 1.04 s (1.41 s p90) | $0.3 | $1.3 | $0.03 | $0.55 | 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.