ruadapt_qwen2.5.3B_ext_u48_instruct_v4: benchmark results
Provider: Other.
Aggregate Rating 1428 ± 35, rank #4739 of 10916 rated models, from 23 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MERA - ruHateSpeech | 74.72 | Accuracy (%) | 51.4 |
| MERA - ruTiE | 72.8 | Accuracy (%) | 51 |
| MERA - ruDetox | 19.68 | Joint Score (%) | 49.8 |
| MERA - RCB | 52.97 | Accuracy (%) | 45.2 |
| MERA - MultiQ | 36.77 | F1 (%) | 43.6 |
| MERA - MaMuRAMu | 63.75 | Accuracy (%) | 42.4 |
| MERA - MathLogicQA | 43.22 | Accuracy (%) | 42.4 |
| MERA - ruWorldTree | 85.9 | Accuracy (%) | 42.4 |
| MERA - ruOpenBookQA | 77 | Accuracy (%) | 42.1 |
| MERA - ruMMLU | 53.28 | Accuracy (%) | 41.4 |
| MERA - SimpleAr | 96.2 | EM (%) | 41.2 |
| MERA - ruHHH | 64.61 | Accuracy (%) | 40.2 |
Interactive version: theaggregate.ai/model?slug=ruadapt-qwen2-5-3b-ext-u48-instruct-v4 · How It Works · Data refreshed daily, snapshot 2026-11-08.