Nova 2.0 Pro Preview: benchmark results
Amazon's Nova 2.0 Pro (Preview), for complex multistep and agentic tasks in their new Nova 2 line, with configurable reasoning-depth levels and a 1M-token context window. Provider: Amazon. Released 2025-12-02. Access: API.
Aggregate Rating 1630 ± 38, rank #357 of 1656 rated models, from 62 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA IFBench | 79.59 | Accuracy (%) | 97.8 |
| AA TAU-2 Bench | 92.69 | Accuracy (%) | 88.3 |
| Tau2-Bench Telecom | 92.69 | Success Rate (%) | 88.1 |
| AA AIME 2025 | 89 | Accuracy (%) | 87.8 |
| AA LiveCodeBench | 73.02 | Pass@1 (%) | 83.9 |
| AA MMLU-Pro | 82.95 | Accuracy (%) | 82.8 |
| AA GPQA Diamond | 78.48 | Accuracy (%) | 64.5 |
| AA Terminal-Bench Hard | 24.24 | Accuracy (%) | 64.5 |
| Artificial Analysis Intelligence Index | 14.16 | Intelligence Index | 54.8 |
| AA Long Context Reasoning | 64.67 | Accuracy (%) | 54.7 |
| AA-Omniscience Hallucination Rate | 79.94 | Hallucination Rate (%) | 50.9 |
| AA Global-MMLU-Lite - Portuguese | 86.75 | Accuracy (%) | 50.5 |
Interactive version: theaggregate.ai/model?slug=nova-2-0-pro-preview · How It Works · Data refreshed daily, snapshot 2026-10-20.