Mistral-Medium-3.5-128B: benchmark results
Provider: Mistral. Access: Open.
Unified ELO 1631 ± 1, rank #180 of 1392 rated models, from 38 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Nejumi 4 - ALT - Bias | 93.33 | Score (%) | 87.7 |
| SEA-HELM (English) | 53.28 | Mean Score | 79.1 |
| Nejumi 4 - GLP - Function Calling | 66.11 | Score (%) | 73.8 |
| SEA-HELM (Filipino) | 66.74 | Mean Score | 67.4 |
| SEA-HELM (Tamil) | 61.37 | Mean Score | 65.1 |
| SEA-HELM (Thai) | 61.13 | Mean Score | 65.1 |
| SEA-HELM (Vietnamese) | 66.33 | Mean Score | 62.8 |
| SEA-HELM (Indonesian) | 67.83 | Mean Score | 60.5 |
| Nejumi 4 - GLP - Application Development | 62.46 | Score (%) | 59.8 |
| SEA-HELM (Malay) | 66.62 | Mean Score | 58.1 |
| SEA-HELM (Burmese) | 32.78 | Mean Score | 55.8 |
| Nejumi 4 - GLP - Knowledge & QA | 67.89 | Score (%) | 54.1 |
Interactive version: theaggregate.ai/model?slug=mistral-medium-3-5-128b · How It Works · Data refreshed daily, snapshot 2026-09-05.