Mistral Medium 3.5 (High): benchmark results
Provider: Mistral. Released 2026-04-29. Access: Open.
Unified ELO 1544 ± 1, rank #656 of 1761 rated models, from 35 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| PACT (Lechmazur) | 1538 | PACT Bilateral Rating | 60 |
| GRIPS | 79.2 | Accuracy (%) | 46.2 |
| GRIPS-hard | 26.2 | Accuracy (%) | 46.2 |
| Vals AI CorpFin v2 | 58.78 | Accuracy (%) | 37.6 |
| Vals AI Finance Agent | 46.11 | Accuracy (%) | 36.7 |
| Vals AI Terminal-Bench 2.0 | 30.34 | Accuracy (%) | 34.1 |
| Vals AI SAGE | 37.61 | Accuracy (%) | 32.9 |
| Vals AI Harvey Legal Agent Bench | 0.42 | Accuracy (%) | 28.1 |
| Vals AI TaxEval v2 | 67.99 | Accuracy (%) | 27.8 |
| ProofBench | 10 | Score (self-reported) | 27.2 |
| MedCode | 33.75 | Score (self-reported) | 23 |
| Vals AI MedCode | 33.75 | Accuracy (%) | 21.3 |
Interactive version: theaggregate.ai/model?slug=mistral-medium-3-5-high · How It Works · Data refreshed daily, snapshot 2026-09-05.