Einstein-v6.1-Llama3-8B: benchmark results
Weyaxi's Einstein v6.1, a full fine-tune of Llama 3 8B on diverse instruction datasets using axolotl, sponsored by sablo.ai. Provider: Weyaxi. Released 2024-04-19. Access: Open.
Aggregate Rating 1416 ± 21, rank #1163 of 1656 rated models, from 21 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - MMLU | 66.19 | Accuracy (%) (5-shot) | 86.2 |
| Open LLM Leaderboard v1 - GSM8K | 66.11 | Accuracy (%) (5-shot) | 81.2 |
| Open Chinese LLM - WinoGrande | 65.04 | Accuracy (%) | 70.7 |
| Open LLM Leaderboard v1 - WinoGrande | 79.32 | Accuracy (%) (5-shot) | 68.6 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 55.1 | MC2 (%) (0-shot) | 64.2 |
| Open LLM Leaderboard - MuSR | 42.13 | Score | 62.1 |
| Open LLM Leaderboard v1 - ARC Challenge | 62.46 | Normalized accuracy (%) (25-shot) | 58 |
| Open LLM Leaderboard v1 - HellaSwag | 82.41 | Normalized accuracy (%) (10-shot) | 51 |
| Open LLM Leaderboard - IFEval | 45.68 | Score | 50.9 |
| Open LLM Leaderboard - BBH | 50.08 | Score | 48.6 |
| Open Chinese LLM - CMMLU | 53.24 | Accuracy (%) | 48.5 |
| Open Chinese LLM - TruthfulQA MC | 53.51 | Accuracy (%) | 48.4 |
Interactive version: theaggregate.ai/model?slug=einstein-v6-1-llama3-8b · How It Works · Data refreshed daily, snapshot 2026-10-20.