cogito-v1-preview-llama-70B: benchmark results
Provider: Nous Research. Released 2025-04-08. Access: Open.
Aggregate Rating 1593 ± 22, rank #1087 of 10916 rated models, from 301 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EuroEval Greek NLU - Greek SA | 84.64 | Matthews correlation coefficient (%) | 100 |
| EuroEval Serbian NLU - MMS SR | 53.61 | Matthews correlation coefficient (%) | 100 |
| MultiWikiQHalluA - Danish | 8 | Answer-level hallucination rate (%): share of the 1,000 Dani | 100 |
| Open Japanese LLM - MMLU EN Exact Match | 86.22 | Score (%) | 100 |
| Open Japanese LLM - HE | 82.55 | Score (%) | 99.8 |
| Open Japanese LLM - Jmmlu Exact Match | 78.88 | Score (%) | 99.8 |
| EuroEval Hungarian NLU - HuSST | 66.96 | Matthews correlation coefficient (%) | 99.5 |
| EuroEval Spanish NLU - Sentiment Headlines ES | 53.48 | Matthews correlation coefficient (%) | 99 |
| EuroEval English | 77.91 | Average Score (%) | 98.8 |
| BenCzechMark - Sentiment | 85.58 | Duel Win Score (%) | 98.7 |
| EuroEval German | 65.67 | Average Score (%) | 98.5 |
| EuroEval Italian | 68.28 | Average Score (%) | 98.1 |
Interactive version: theaggregate.ai/model?slug=cogito-v1-preview-llama-70b · How It Works · Data refreshed daily, snapshot 2026-11-08.