SOLAR-Platypus-10.7B-v2: benchmark results
SOLAR fine-tune on Open-Platypus with rank-64 QLoRA covering MLP and attention projections and a Human/Assistant template. Provider: kyujinpy. Access: Open.
Aggregate Rating 1381 ± 37, rank #6802 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - WinoGrande | 81.45 | Accuracy (%) (5-shot) | 79.3 |
| Open LLM Leaderboard v1 - HellaSwag | 83.57 | Normalized accuracy (%) (10-shot) | 59.8 |
| Open LLM Leaderboard v1 - MMLU | 59.93 | Accuracy (%) (5-shot) | 46.3 |
| Open LLM Leaderboard v1 - ARC Challenge | 59.39 | Normalized accuracy (%) (25-shot) | 43.2 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 43.15 | MC2 (%) (0-shot) | 22.7 |
| Open LLM Leaderboard v1 - GSM8K | 4.02 | Accuracy (%) (5-shot) | 22.3 |
Interactive version: theaggregate.ai/model?slug=solar-platypus-10-7b-v2 · How It Works · Data refreshed daily, snapshot 2026-11-08.