cogito-v1-preview-qwen-32B: benchmark results
Provider: Nous Research. Released 2025-04-08. Access: Open.
Aggregate Rating 1577 ± 20, rank #1297 of 10916 rated models, from 330 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MultiWikiQHalluA - English | 1 | Answer-level hallucination rate (%): share of the 1,000 Engl | 100 |
| MultiWikiQHalluA - Icelandic | 18 | Answer-level hallucination rate (%): share of the 1,000 Icel | 100 |
| Open Japanese LLM - HE | 83.35 | Score (%) | 100 |
| Open Japanese LLM - Jmmlu Exact Match | 80.51 | Score (%) | 100 |
| Open Japanese LLM - MMLU EN Exact Match | 86.19 | Score (%) | 99.8 |
| Open Japanese LLM - MR | 95 | Score (%) | 99.6 |
| Open Japanese LLM - Jcommonsenseqa Exact Match | 96.87 | Score (%) | 98.9 |
| EuroEval Portuguese NLU - MultiWikiQA PT | 79.44 | Character F1 (%) | 98.8 |
| EuroEval Romanian NLU - RoSent | 97.85 | Matthews correlation coefficient (%) | 98.5 |
| EuroEval Spanish NLU - Sentiment Headlines ES | 53.21 | Matthews correlation coefficient (%) | 98.5 |
| EuroEval English Common Sense Reasoning | 90.5 | Common Sense Reasoning Average Score (%) | 98.4 |
| EuroEval French Common Sense Reasoning | 82.35 | Common Sense Reasoning Average Score (%) | 98 |
Interactive version: theaggregate.ai/model?slug=cogito-v1-preview-qwen-32b · How It Works · Data refreshed daily, snapshot 2026-11-08.