BerryLM-L-v2-early-ckpt: benchmark results
Provider: Other.
Aggregate Rating 1743 ± 39, rank #204 of 10916 rated models, from 23 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MERA - ruCodeEval | 84.76 | pass@1 (%) | 100 |
| MERA - MathLogicQA | 99.83 | Accuracy (%) | 99.8 |
| MERA - ruHHH | 93.82 | Accuracy (%) | 99.8 |
| MERA - ruHumanEval | 82.87 | pass@1 (%) | 99.5 |
| MERA | 0.82 | Total score | 98.6 |
| MERA - ruWorldTree | 99.62 | Accuracy (%) | 98.6 |
| MERA - BPS | 100 | Accuracy (%) | 98.3 |
| MERA - ruOpenBookQA | 96.75 | Accuracy (%) | 98.3 |
| MERA - USE | 72.16 | Grade, normalized (%) | 98.1 |
| MERA - MaMuRAMu | 90.37 | Accuracy (%) | 97.6 |
| MERA - RWSD | 88.85 | Accuracy (%) | 97.6 |
| MERA - ruModAr | 99.9 | EM (%) | 97.2 |
Interactive version: theaggregate.ai/model?slug=berrylm-l-v2-early-ckpt · How It Works · Data refreshed daily, snapshot 2026-11-08.