Llama 3 8B Instruct 262k: benchmark results
Provider: Meta. Released 2024-04-18. Access: Open.
Unified ELO 1439 ± 26, rank #1036 of 1629 rated models, from 29 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Medical LLM - MedMCQA | 57.4 | Accuracy (%) | 82.1 |
| Open Medical LLM - MMLU Medical Genetics | 81 | Accuracy (%) | 71.6 |
| Open Portuguese LLM - OAB Exams | 46.74 | Accuracy (%) | 64.4 |
| Open Medical LLM - MMLU Anatomy | 63.7 | Accuracy (%) | 63.4 |
| Open Medical LLM | 67.67 | Average Accuracy (%) | 59.1 |
| Open Medical-LLM Leaderboard | 67.67 | Medical average (%) | 57.8 |
| Open Medical LLM - MedQA (USMLE) | 57.11 | Accuracy (%) | 56.2 |
| Open Medical LLM - MMLU College Biology | 75 | Accuracy (%) | 56 |
| Open Medical LLM - MMLU Professional Medicine | 70.22 | Accuracy (%) | 52.8 |
| Open Portuguese LLM - ENEM | 61.58 | Accuracy (%) | 49.9 |
| Open Medical LLM - MMLU Clinical Knowledge | 70.57 | Accuracy (%) | 48.3 |
| Open Portuguese LLM - BLUEX | 50.76 | Accuracy (%) | 48.1 |
Interactive version: theaggregate.ai/model?slug=llama-3-8b-instruct-262k · How It Works · Data refreshed daily, snapshot 2026-10-07.