INTELLECT-1-Instruct: benchmark results
Provider: PrimeIntellect. Released 2024-09-01. Access: Open.
Unified ELO 1322 ± 1, rank #1320 of 1392 rated models, from 12 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| ARC Challenge (AI2) | 54.52 | Accuracy (%) | 48.7 |
| GSM8K | 38.58 | Accuracy (%) | 44.1 |
| HellaSwag | 71.42 | Accuracy (%) | 35.5 |
| WinoGrande | 65.82 | Accuracy (%) | 26.2 |
| MMLU | 49.89 | Accuracy (%) | 20 |
| Open LLM Leaderboard - MuSR | 3.71 | Score | 17.8 |
| Open LLM Leaderboard - MATH Level 5 | 2.27 | Score | 15.4 |
| Big-Bench Hard | 34.85 | Average (%) | 10.4 |
| Open LLM Leaderboard - GPQA | 0 | Score | 2.6 |
| Open LLM Leaderboard - BBH | 1.75 | Score | 1.2 |
| Open LLM Leaderboard - IFEval | 0 | Score | 0.1 |
| Open LLM Leaderboard - MMLU-Pro | 0.71 | Score | 0.1 |
Interactive version: theaggregate.ai/model?slug=intellect-1-instruct · How It Works · Data refreshed daily, snapshot 2026-09-05.