Mistral-10.7B-Instruct-v0.3-depth-upscaling: benchmark results
Provider: Mistral. Access: Open.
Aggregate Rating 1406 ± 33, rank #5794 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 60.65 | MC2 (%) (0-shot) | 76.3 |
| Open LLM Leaderboard v1 - WinoGrande | 78.45 | Accuracy (%) (5-shot) | 63.3 |
| Open LLM Leaderboard v1 - ARC Challenge | 63.14 | Normalized accuracy (%) (25-shot) | 61.5 |
| Open LLM Leaderboard v1 - HellaSwag | 83.29 | Normalized accuracy (%) (10-shot) | 57.6 |
| Open LLM Leaderboard v1 - MMLU | 62.31 | Accuracy (%) (5-shot) | 55 |
| Open LLM Leaderboard v1 - GSM8K | 36.39 | Accuracy (%) (5-shot) | 51.8 |
Interactive version: theaggregate.ai/model?slug=mistral-10-7b-instruct-v0-3-depth-upscaling · How It Works · Data refreshed daily, snapshot 2026-11-08.