Mixtral_AI_Cyber_3.0: benchmark results
Provider: Mistral.
Aggregate Rating 1409 ± 33, rank #5661 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - WinoGrande | 80.19 | Accuracy (%) (5-shot) | 72.8 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 58.21 | MC2 (%) (0-shot) | 71.9 |
| Open LLM Leaderboard v1 - HellaSwag | 84.02 | Normalized accuracy (%) (10-shot) | 63.2 |
| Open LLM Leaderboard v1 - GSM8K | 45.94 | Accuracy (%) (5-shot) | 59.7 |
| Open LLM Leaderboard v1 - ARC Challenge | 62.46 | Normalized accuracy (%) (25-shot) | 58 |
| Open LLM Leaderboard v1 - MMLU | 61.91 | Accuracy (%) (5-shot) | 53.3 |
Interactive version: theaggregate.ai/model?slug=mixtral-ai-cyber-3-0 · How It Works · Data refreshed daily, snapshot 2026-11-08.