Llama 3 8B Instruct Coder: benchmark results
A Llama-3-8B-Instruct coding fine-tune by rombodawg using combined CodeFeedback instruction data and Qalore training. Provider: rombodawg. Access: Open.
Aggregate Rating 1414 ± 35, rank #5453 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - MMLU | 66.35 | Accuracy (%) (5-shot) | 86.9 |
| Open LLM Leaderboard v1 - GSM8K | 67.32 | Accuracy (%) (5-shot) | 83.4 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 52.33 | MC2 (%) (0-shot) | 55.4 |
| Open LLM Leaderboard v1 - ARC Challenge | 60.67 | Normalized accuracy (%) (25-shot) | 48.9 |
| Open LLM Leaderboard v1 - WinoGrande | 74.9 | Accuracy (%) (5-shot) | 39.3 |
| Open LLM Leaderboard v1 - HellaSwag | 77.99 | Normalized accuracy (%) (10-shot) | 32.6 |
Interactive version: theaggregate.ai/model?slug=llama-3-8b-instruct-coder · How It Works · Data refreshed daily, snapshot 2026-11-08.