Llama 3 8B Ee: benchmark results
BEE-spoke-data release: Continued-pretraining derivative of Meta-Llama-3-8B on the publisher's bees-internal dataset; it remains distinct from the original Meta checkpoint. Provider: BEE-spoke-data. Access: Open.
Aggregate Rating 1402 ± 31, rank #5980 of 10916 rated models, from 12 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - MMLU | 66.01 | Accuracy (%) (5-shot) | 85.7 |
| Open LLM Leaderboard - GPQA | 31.38 | Score | 70.6 |
| Open LLM Leaderboard v1 - GSM8K | 39.88 | Accuracy (%) (5-shot) | 55 |
| Open LLM Leaderboard v1 - WinoGrande | 77.11 | Accuracy (%) (5-shot) | 53.3 |
| Open LLM Leaderboard v1 - HellaSwag | 81.48 | Normalized accuracy (%) (10-shot) | 45.2 |
| Open LLM Leaderboard - MMLU-Pro | 32.2 | Score | 44.2 |
| Open LLM Leaderboard v1 - ARC Challenge | 58.62 | Normalized accuracy (%) (25-shot) | 40.4 |
| Open LLM Leaderboard - BBH | 46.26 | Score | 36.9 |
| Open LLM Leaderboard - MATH Level 5 | 4.83 | Score | 27.2 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 42.9 | MC2 (%) (0-shot) | 22 |
| Open LLM Leaderboard - MuSR | 36.54 | Score | 20.7 |
| Open LLM Leaderboard - IFEval | 19.51 | Score | 11.2 |
Interactive version: theaggregate.ai/model?slug=llama-3-8b-ee · How It Works · Data refreshed daily, snapshot 2026-11-08.