Nova Premier: benchmark results
Amazon's Nova Premier, the most capable model of the first Nova generation (December 2024). Provider: Amazon. Released 2024-12-03. Access: API.
Unified ELO 1548 ± 1, rank #448 of 1392 rated models, from 62 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| HELM Long Context - InfiniteBench En.MC | 92 | EM | 90 |
| AA Omniscience - Software Engineering (SWE) - Dart | 28 | Accuracy (%) | 82.1 |
| Wolfram LLM Benchmarking Project | 55.7 | Correct Functionality (%) | 80.9 |
| MMMU Benchmark | 68 | Validation Score | 75.1 |
| HELM Long Context - RULER HotPotQA | 57 | RULER String Match | 75 |
| AA Omniscience - Software Engineering (SWE) - Python | 26 | Accuracy (%) | 74.4 |
| AA Omniscience - Software Engineering (SWE) - Kotlin | 24 | Accuracy (%) | 72.5 |
| AA Omniscience - Software Engineering (SWE) - Julia | 16 | Accuracy (%) | 68.8 |
| AA Omniscience - Software Engineering (SWE) - PHP | 26 | Accuracy (%) | 65 |
| AA Omniscience - Health | 24.6 | Accuracy (%) | 64.8 |
| AA Omniscience - Software Engineering (SWE) - TypeScript | 23.33 | Accuracy (%) | 62.1 |
| AA Omniscience - Software Engineering (SWE) - HTML | 32 | Accuracy (%) | 60.4 |
Interactive version: theaggregate.ai/model?slug=nova-premier · How It Works · Data refreshed daily, snapshot 2026-09-05.