Mistral Large 4: benchmark results
Provider: Mistral. Access: Open.
Unified ELO 1744 ± 28, rank #121 of 1629 rated models, from 27 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Vals AI Harvey Legal Agent Bench | 15.83 | Accuracy (%) | 92.6 |
| Vals AI Vibe Code Bench | 78.4 | Accuracy (%) | 77.1 |
| Vals AI Finance Agent v2 | 54.68 | Accuracy (%) | 70.7 |
| BenchLM | 53.7 | Overall Score | 66.2 |
| Vals AI Tax Agent Bench | 63.29 | Accuracy (%) | 63.1 |
| LM Market Cap LMC Score | 68.1 | LMC Score (0-100) | 62.3 |
| PLCC - History | 83 | Accuracy (%) | 58.1 |
| Vals AI Terminal-Bench 4.0 | 22.73 | Accuracy (%) | 57 |
| PLCC - Culture & Tradition | 73 | Accuracy (%) | 54.7 |
| PLCC - Vocabulary | 63 | Accuracy (%) | 53.2 |
| Vals AI Code Migration | 30.56 | Accuracy (%) | 49.3 |
| Vals AI Legal Research Bench | 31.73 | Accuracy (%) | 49.3 |
Interactive version: theaggregate.ai/model?slug=mistral-large-4 · How It Works · Data refreshed daily, snapshot 2026-10-07.