Nova Micro: benchmark results
Amazon's Nova Micro, the smallest, text-only, lowest-cost first-generation Nova model (December 2024). Provider: Amazon. Released 2024-12-03. Access: API.
Unified ELO 1485 ± 1, rank #759 of 1392 rated models, from 68 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| OpenRouter Tau2-Bench Airline | 78.7 | Accuracy (%) | 95.8 |
| HELM NaturalQuestions (Open) | 77.86 | F1 (%) | 91.1 |
| OpenRouter GPQA Diamond | 89.1 | Accuracy (%) | 89.7 |
| HELM (Stanford) | 52.4 | Mean Win Rate (%) | 55.6 |
| LLM Stats (DROP) | 79.3 | Score (%) | 53.6 |
| HELM NarrativeQA | 74.38 | F1 (%) | 53.3 |
| HELM WMT 2014 | 19.2 | BLEU-4 (%) | 51.1 |
| Translation en to Set1 spBleu | 40.2 | en→Set1 spBLEU (self-reported) | 50 |
| Translation Set1 to en COMET22 | 88.7 | Set1→en COMET22 (self-reported) | 45.8 |
| Translation Set1 to en spBleu | 42.6 | Set1→en spBLEU (self-reported) | 41.7 |
| Translation en to Set1 COMET22 | 88.5 | en→Set1 COMET22 (self-reported) | 41.7 |
| AA Omniscience | -48.52 | Score | 36.4 |
Interactive version: theaggregate.ai/model?slug=nova-micro · How It Works · Data refreshed daily, snapshot 2026-09-05.