falcon-7B Instruct: benchmark results
Provider: TII. Released 2023-05-25. Access: Open.
Unified ELO 1258 ± 1, rank #1382 of 1392 rated models, from 98 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| LLM Trustworthy - Fairness | 100 | Trust Score (%) | 94 |
| MMLU-by-task - TruthfulQA MC1 | 28.89 | Accuracy (%) | 52 |
| MMLU-by-task - Machine Learning | 32.14 | Accuracy (%) | 48.7 |
| MMLU-by-task - TruthfulQA MC2 | 44.07 | Accuracy (%) | 46.6 |
| LLM Trustworthy - Stereotype | 87 | Trust Score (%) | 46 |
| MMLU-by-task - Moral Scenarios | 25.14 | Accuracy (%) | 45.4 |
| MMLU-by-task - Abstract Algebra | 29 | Accuracy (%) | 45.3 |
| LLM Trustworthy - Ethics | 50.28 | Trust Score (%) | 44 |
| MMLU-by-task - ARC Challenge | 42.15 | Accuracy (%) | 36.7 |
| LLM Trustworthy - Adversarial | 43.98 | Trust Score (%) | 36 |
| LLM Trustworthy - Toxicity | 39 | Trust Score (%) | 36 |
| MMLU-by-task - Formal Logic | 26.98 | Accuracy (%) | 35.6 |
Interactive version: theaggregate.ai/model?slug=falcon-7b-instruct · How It Works · Data refreshed daily, snapshot 2026-09-05.