Nova Premier
AI model by Amazon. Real-time pricing and benchmark data.
Benchmarks
Editorial Analysis
Profile. Nova Premier is positioned in the mid-tier at $12.50/M output tokens. The benchmark profile is balanced — scoring across standard tests — making it a default choice for teams that need general-purpose reasoning without paying frontier-model prices.
Latency. On latency, Nova Premier reports 33 tok/s output speed and 2770ms time-to-first-token. Output speed is on the slower end; for real-time chat consider pairing with a streaming front-end or routing latency-sensitive requests to a faster model.
Cost model. For a workload of 1M input tokens + 100K output tokens, Nova Premier costs $3.75 per million input-equivalent requests. Multiply by your monthly request volume to project the bill, and compare against the speed/correctness profile above to decide if the trade-off is worth it for your workload.
Editorial summary generated from public pricing and benchmark data (Artificial Analysis API). All numbers are sourced — see the Artificial Analysis methodology for benchmark definitions. Last refreshed 2026-08-24.
Compare with similar models
| Model | Input | Output | Speed |
|---|---|---|---|
Nova PremierCurrent | $2.50 | $12.50 | 33 tok/s |
| GPT-5.6 Terra (medium) | $2.00 | $12.00 | 98 tok/s |
| Mistral Large (Feb '24) | $4.00 | $12.00 | — |
| GPT-5.6 Terra (high) | $2.00 | $12.00 | 103 tok/s |
| GPT-5.6 Terra (Non-reasoning) | $2.00 | $12.00 | 99 tok/s |
| GPT-5.6 Terra (xhigh) | $2.00 | $12.00 | 101 tok/s |