AmazonAPI Host Benchmarks
Nova 2 0 Pro Reasoning Low — API Providers & Speed Matrix
Compare live throughput, TTFT latency distributions, prompt cache pricing, and hardware stacks across cloud hosting endpoints.
Fastest Provider
500 t/s
Lowest Latency (TTFT)
0.09s P50
Best Price Value
$0.36 / $1.6
Blended Pricing Workload Calculator
Switch between standard conversational ratios, agentic caching loops, and heavy RAG workloads.
| Provider & Host | Hardware Stack | Output Speed (TPS) | Latency TTFT (P50) | Input Price ($/1M) | Output Price ($/1M) | Cache Read ($/1M) | Blended Price ($/1M) | SLA |
|---|---|---|---|---|---|---|---|---|
AmazonOfficial | Cloud TPU / Nvidia H100 | 40 t/s | 0.8s (1.12s p90) | $0.8 | $3.2 | $0.08 | $1.4 | 99.99% |
Groq Cloud | Groq LPU v2 (SRAM) | 168 t/s | 0.09s (0.12s p90) | $0.68 | $2.72 | $0.064 | $1.19 | 99.95% |
Cerebras Cloud | CS-3 Wafer-Scale Engine | 500 t/s | 0.12s (0.16s p90) | $0.72 | $2.88 | — | $1.26 | 99.90% |
Together AI | Nvidia H100 SXM5 | 60 t/s | 0.64s (0.88s p90) | $0.52 | $2.24 | $0.048 | $0.95 | 99.95% |
DeepInfra | Nvidia H100 / L40S | 52 t/s | 0.72s (1s p90) | $0.36 | $1.6 | $0.04 | $0.67 | 99.90% |
Fireworks AI | Nvidia B200 / H100 | 76 t/s | 0.56s (0.76s p90) | $0.48 | $2.08 | $0.048 | $0.88 | 99.99% |