dolphin-2.8-experiment26-7B [float16] [rev: b7d94074abb2]: benchmark results
Provider: Cognitive Computations. Access: Open.
Aggregate Rating 1424 ± 34, rank #4995 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - WinoGrande | 81.61 | Accuracy (%) (5-shot) | 80.1 |
| Open LLM Leaderboard v1 - GSM8K | 63.61 | Accuracy (%) (5-shot) | 77.2 |
| Open LLM Leaderboard v1 - ARC Challenge | 64.51 | Normalized accuracy (%) (25-shot) | 66.8 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 54.87 | MC2 (%) (0-shot) | 63.7 |
| Open LLM Leaderboard v1 - HellaSwag | 83.79 | Normalized accuracy (%) (10-shot) | 61.5 |
| Open LLM Leaderboard v1 - MMLU | 63.24 | Accuracy (%) (5-shot) | 59.3 |
Interactive version: theaggregate.ai/model?slug=dolphin-2-8-experiment26-7b-float16-rev-b7d94074abb2 · How It Works · Data refreshed daily, snapshot 2026-11-08.