Nova 2.0 Lite (Non-reasoning): benchmark results
Provider: Amazon. Released 2025-12-02. Access: API.
Aggregate Rating 1485 ± 11, rank #1150 of 2201 rated models, from 257 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Ko-AgentBench - L1 Single Tool Call | 79.55 | Argument Accuracy (%) | 100 |
| AGC-Bench - grapheval_ai_researcher | 1.11 | Dataset z-score | 96.3 |
| Nejumi 4 - Toxicity - Violation Categories | 57.14 | Criteria met (%) | 95.5 |
| AGC-Bench - hypobench | 0.42 | Dataset z-score | 93.9 |
| Vectara Hallucination Leaderboard | 94.9 | Factual Consistency Rate (%) | 90 |
| AGC-Bench - mops | 1.05 | Dataset z-score | 89 |
| Context Arena MRCR (2-needle) | 63.5 | AUC@1M (%) | 87.5 |
| Ko-AgentBench - L5 Error Handling & Robustness | 26.25 | Adaptive Routing Score (%) | 85.7 |
| Nejumi 4 - BFCL - Relevance Detection | 83.33 | Accuracy (%) | 82.5 |
| Ko-AgentBench - L3 Sequential Tool Reasoning | 100 | PSM (%) | 82.1 |
| AGC-Bench - tinystories | 0.72 | Dataset z-score | 79.3 |
| Context Arena MRCR (4-needle) | 47.5 | AUC@1M (%) | 78 |
Interactive version: theaggregate.ai/model?slug=nova-2-0-lite-non-reasoning · How It Works · Data refreshed daily, snapshot 2026-11-12.