WestLake-7B-v2: benchmark results
Community creator senseable's Mistral-based WestLake v2, a 7B tune aimed at role-play consistency and creative writing. Provider: senseable. Released 2024-01-22. Access: Open.
Aggregate Rating 1474 ± 31, rank #871 of 1656 rated models, from 33 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - WinoGrande | 86.98 | Accuracy (%) (5-shot) | 99.4 |
| Open LLM Leaderboard v1 - ARC Challenge | 73.04 | Normalized accuracy (%) (25-shot) | 95.2 |
| Open LLM Leaderboard v1 - HellaSwag | 88.65 | Normalized accuracy (%) (10-shot) | 93.4 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 67.06 | MC2 (%) (0-shot) | 85.4 |
| Open LLM Leaderboard v1 - GSM8K | 67.63 | Accuracy (%) (5-shot) | 84 |
| Open LLM Leaderboard v1 - MMLU | 64.71 | Accuracy (%) (5-shot) | 75.3 |
| EuroEval Spanish NLU - Sentiment Headlines ES | 45.49 | Matthews correlation coefficient (%) | 71.4 |
| EuroEval Italian NLU - MultiNERD IT | 69.6 | Micro F1 excluding MISC (%) | 70.7 |
| EuroEval Italian NLU - SQuAD IT | 69.38 | Character F1 (%) | 65.1 |
| EuroEval Norwegian Knowledge | 27.91 | Knowledge Average Score (%) | 64.8 |
| EuroEval Italian NLU | 52.92 | NLU Average Score (%) | 63.7 |
| EuroEval Italian NLU - Sentipolc16 | 54.85 | Matthews correlation coefficient (%) | 62.4 |
Interactive version: theaggregate.ai/model?slug=westlake-7b-v2 · How It Works · Data refreshed daily, snapshot 2026-10-20.