Mistral-Instruct-7B-v0.2-ChatAlpacaV2-4bit [4bit]: benchmark results
Provider: Mistral. Access: Open.
Aggregate Rating 1399 ± 34, rank #6155 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 67.29 | MC2 (%) (0-shot) | 85.7 |
| Open LLM Leaderboard v1 - HellaSwag | 84.55 | Normalized accuracy (%) (10-shot) | 66.8 |
| Open LLM Leaderboard v1 - ARC Challenge | 62.12 | Normalized accuracy (%) (25-shot) | 56.1 |
| Open LLM Leaderboard v1 - GSM8K | 40.33 | Accuracy (%) (5-shot) | 55.6 |
| Open LLM Leaderboard v1 - WinoGrande | 77.11 | Accuracy (%) (5-shot) | 53.3 |
| Open LLM Leaderboard v1 - MMLU | 60.66 | Accuracy (%) (5-shot) | 49.1 |
Interactive version: theaggregate.ai/model?slug=mistral-instruct-7b-v0-2-chatalpacav2-4bit-4bit · How It Works · Data refreshed daily, snapshot 2026-11-08.