Zarya-4B: benchmark results
Provider: Other. Released 2026-09-18. Access: Open.
Aggregate Rating 1435 ± 38, rank #4590 of 10552 rated models, from 17 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MERA v2 - SAGE | 48.6 | Score (%) | 30.6 |
| MERA v2 - Culture Specific | 16.6 | Score (%) | 20.4 |
| MERA v2 - RussianRegions | 3.6 | Score (%) | 20.4 |
| MERA v2 - Characters | 27.2 | Score (%) | 16.3 |
| MERA v2 - GorillaHard | 4.8 | Score (%) | 12.2 |
| MERA v2 - RUBIN | 26 | Score (%) | 12.2 |
| MERA v2 - Agentic | 2.5 | Score (%) | 11.2 |
| MERA v2 - Riddles | 5 | Score (%) | 11.2 |
| MERA v2 | 0.08 | Total score | 10.2 |
| MERA v2 - Human Centric | 7.4 | Score (%) | 10.2 |
| MERA v2 - IFHardBench | 3.2 | Score (%) | 10.2 |
| MERA v2 - LIMUR | 30.4 | Score (%) | 10.2 |
Interactive version: theaggregate.ai/model?slug=zarya-4b · How It Works · Data refreshed daily, snapshot 2026-11-16.