DolphinStar-12.5B: benchmark results
Provider: Cognitive Computations. Access: Open.
Aggregate Rating 1387 ± 32, rank #6647 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 51.51 | MC2 (%) (0-shot) | 52.2 |
| Open LLM Leaderboard v1 - GSM8K | 35.41 | Accuracy (%) (5-shot) | 51.1 |
| Open LLM Leaderboard v1 - ARC Challenge | 60.84 | Normalized accuracy (%) (25-shot) | 49.8 |
| Open LLM Leaderboard v1 - HellaSwag | 82 | Normalized accuracy (%) (10-shot) | 48.3 |
| Open LLM Leaderboard v1 - MMLU | 60.36 | Accuracy (%) (5-shot) | 47.9 |
| Open LLM Leaderboard v1 - WinoGrande | 76.24 | Accuracy (%) (5-shot) | 47 |
Interactive version: theaggregate.ai/model?slug=dolphinstar-12-5b · How It Works · Data refreshed daily, snapshot 2026-11-08.