dolphin-2.6-mixtral-8x7B: benchmark results
Provider: Cognitive Computations. Released 2023-12-21. Access: Open.
Aggregate Rating 1357 ± 47, rank #7524 of 10916 rated models, from 9 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EvalPlus (HumanEval+ & MBPP+) | 58.1 | Pass@1 avg (%) | 54.8 |
| Open Chinese LLM - CMMLU | 50.47 | Accuracy (%) | 36.7 |
| Open Chinese LLM - C-Eval Semantic | 58.56 | Accuracy (%) | 19.8 |
| Open Chinese LLM - GSM8K | 23.88 | Accuracy (%) | 17.3 |
| Open Chinese LLM - WinoGrande | 59.98 | Accuracy (%) | 13.6 |
| Open Chinese LLM - TruthfulQA MC | 47.72 | Accuracy (%) | 10.9 |
| Open Chinese LLM Leaderboard | 45.69 | Official Nonzero-Task Mean (%) | 9.5 |
| Open Chinese LLM - ARC Challenge | 41.04 | Accuracy (%) | 8.3 |
| Open Chinese LLM - HellaSwag | 49.17 | Accuracy (%) | 7.1 |
Interactive version: theaggregate.ai/model?slug=dolphin-2-6-mixtral-8x7b · How It Works · Data refreshed daily, snapshot 2026-11-08.