Sheared-LLaMA-1.3B: benchmark results
Provider: Princeton NLP. Released 2023-10-10. Access: Open.
Aggregate Rating 1128 ± 26, rank #10366 of 10916 rated models, from 13 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| PIQA (PapersWithCode) | 73.4 | Accuracy | 26.2 |
| Open LLM Leaderboard - MuSR | 37.13 | Score | 24.7 |
| Open LLM Leaderboard - IFEval | 21.98 | Score | 14.8 |
| Open LLM Leaderboard v1 - HellaSwag | 60.91 | Normalized accuracy (%) (10-shot) | 14.8 |
| Open LLM Leaderboard v1 - GSM8K | 0.45 | Accuracy (%) (5-shot) | 12.3 |
| Open LLM Leaderboard - BBH | 31.97 | Score | 11.7 |
| Open LLM Leaderboard v1 - WinoGrande | 58.64 | Accuracy (%) (5-shot) | 11.7 |
| Open LLM Leaderboard v1 - ARC Challenge | 32.85 | Normalized accuracy (%) (25-shot) | 11.2 |
| Open LLM Leaderboard v1 - MMLU | 25.71 | Accuracy (%) (5-shot) | 9.5 |
| Open LLM Leaderboard - MATH Level 5 | 1.28 | Score | 9.2 |
| Open LLM Leaderboard - MMLU-Pro | 11.71 | Score | 8.5 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 37.14 | MC2 (%) (0-shot) | 5.1 |
Interactive version: theaggregate.ai/model?slug=sheared-llama-1-3b · How It Works · Data refreshed daily, snapshot 2026-11-08.