GigaChat 3.5 Ultra: benchmark results
Provider: Other.
Aggregate Rating 1606 ± 34, rank #959 of 10916 rated models, from 17 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MERA v2 - SAGE | 72.2 | Score (%) | 92.4 |
| MERA v2 - Culture Specific | 38.3 | Score (%) | 80.4 |
| MERA v2 - RUBIN | 67.6 | Score (%) | 78.3 |
| MERA v2 - RussianRegions | 11.5 | Score (%) | 77.2 |
| MERA v2 - Riddles | 45.4 | Score (%) | 69.6 |
| MERA v2 - Characters | 56.9 | Score (%) | 67.4 |
| MERA v2 - Human Centric | 36.8 | Score (%) | 52.2 |
| MERA v2 | 0.25 | Total score | 45.7 |
| MERA v2 - Enantiosemy | 37.2 | Score (%) | 43.5 |
| MERA v2 - Humor | 19.3 | Score (%) | 43.5 |
| MERA v2 - LIMUR | 53.9 | Score (%) | 43.5 |
| MERA v2 - NewReasoning | 39.4 | Score (%) | 41.3 |
Interactive version: theaggregate.ai/model?slug=gigachat-3-5-ultra · How It Works · Data refreshed daily, snapshot 2026-11-08.