bellman-7B-mistral-instruct-v0.2: benchmark results
Provider: neph1. Released 2023-12-16. Access: Open.
Aggregate Rating 1374 ± 33, rank #7059 of 10916 rated models, from 27 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EuroEval Danish NLU - Angry Tweets | 47.58 | Matthews correlation coefficient (%) | 52 |
| EuroEval Swedish NLU - SUC3 | 54.38 | Micro F1 excluding MISC (%) | 48.4 |
| EuroEval Danish Knowledge | 43.64 | Knowledge Average Score (%) | 45.1 |
| EuroEval Danish NLU - Dansk | 46.11 | Micro F1 excluding MISC (%) | 41.8 |
| EuroEval Danish NLU - ScaLA DA | 18.41 | Matthews correlation coefficient (%) | 41.5 |
| EuroEval Norwegian NLU - NorNE NB | 57.01 | Micro F1 excluding MISC (%) | 41.3 |
| EuroEval Norwegian NLU - Norec | 38.81 | Matthews correlation coefficient (%) | 40.9 |
| EuroEval Danish NLU | 37.37 | NLU Average Score (%) | 40.7 |
| EuroEval Norwegian NLU - NorNE NN | 56.77 | Micro F1 excluding MISC (%) | 40.2 |
| EuroEval Danish Knowledge - Danish Citizen Tests | 43.64 | MCC (x100) | 39.5 |
| EuroEval Norwegian NLU - ScaLA NN | 9.29 | Matthews correlation coefficient (%) | 38.9 |
| EuroEval Danish | 33.47 | Average Score (%) | 38.7 |
Interactive version: theaggregate.ai/model?slug=bellman-7b-mistral-instruct-v0-2 · How It Works · Data refreshed daily, snapshot 2026-11-08.