Llama 3.2 1B: benchmark results
Meta's Llama 3.2 1B, the base pretrained checkpoint in Meta's Llama 3.2 family released September 2024; the linked model card provides no further details here. Provider: Meta. Released 2024-09-25. Access: Open.
Aggregate Rating 1233 ± 10, rank #1561 of 1656 rated models, from 677 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Polar (US) - Sociocultural Axis | 87.31 | Sociocultural-axis ICAT (0-100), the StereoSet idealized con | 100 |
| aiXamine - Model & Data Privacy | 86.2 | Service score (0-100; normalized pass rate) | 95.1 |
| Polar (US) | 79.27 | Total ICAT (0-100), the StereoSet idealized context associat | 94.6 |
| Polar (US) - Economic Axis | 71.22 | Economic-axis ICAT (0-100), the StereoSet idealized context | 77 |
| LA Leaderboard - RagQuAS | 78.96 | Semantic Answer Similarity (x100) | 76.5 |
| LA Leaderboard - MGSM Spanish | 2.4 | Exact Match (%) | 73.1 |
| LA Leaderboard - GalCoLA | 6.29 | MCC (x100) | 70.6 |
| LA Leaderboard - HumorQA | 58.82 | Accuracy (%) | 70.6 |
| aiXamine - Safety Alignment | 97 | Service score (0-100; normalized pass rate) | 65.9 |
| LA Leaderboard - MGSM Catalan | 4 | Exact Match (%) | 64.6 |
| aiXamine - Jailbreak Robustness | 67.6 | Service score (0-100; normalized pass rate) | 63.4 |
| Open Japanese LLM - Mbpp Pylint Check | 59.64 | Score (%) | 60.6 |
Interactive version: theaggregate.ai/model?slug=llama-3-2-1b · How It Works · Data refreshed daily, snapshot 2026-10-20.