Cotype Light 3: benchmark results
Provider: MTS AI.
Aggregate Rating 1661 ± 43, rank #551 of 10916 rated models, from 24 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MERA - MultiQ | 75.14 | F1 (%) | 99.1 |
| MERA | 0.79 | Total score | 96 |
| MERA - MathLogicQA | 99.48 | Accuracy (%) | 96 |
| MERA - LCS | 71 | Accuracy (%) | 95.7 |
| MERA - SimpleAr | 100 | EM (%) | 95.5 |
| MERA - ruWorldTree | 99.43 | Accuracy (%) | 95.5 |
| MERA - ruMultiAr | 99.8 | EM (%) | 94.8 |
| MERA - RCB | 60.05 | Accuracy (%) | 92.9 |
| MERA - RWSD | 79.62 | Accuracy (%) | 92.9 |
| MERA - ruHumanEval | 64.88 | pass@1 (%) | 92.8 |
| MERA - ruCodeEval | 66.22 | pass@1 (%) | 92.4 |
| MERA - USE | 59.12 | Grade, normalized (%) | 91 |
Interactive version: theaggregate.ai/model?slug=cotype-light-3 · How It Works · Data refreshed daily, snapshot 2026-11-08.