Xwin-LM-7B-V0.2: benchmark results
Provider: Other. Access: Open.
Unified ELO 1315 ± 17, rank #2403 of 2656 rated models, from 19 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - KorNAT-Harmless | 64.43 | Normalized accuracy (%) | 74.3 |
| Open Korean LLM - Ko-GPQA-Diamond | 23.23 | Normalized accuracy (%) | 48.1 |
| Open Portuguese LLM - FaQuAD NLI | 56.75 | Macro F1 (%) | 39.5 |
| Open Korean LLM - KorNAT-Helpful | 44.8 | Normalized accuracy (%) | 32.8 |
| Open Portuguese LLM - OAB Exams | 34.53 | Accuracy (%) | 31.3 |
| Open Portuguese LLM - ENEM | 42.9 | Accuracy (%) | 31.2 |
| Open Portuguese LLM - BLUEX | 37.41 | Accuracy (%) | 30.8 |
| Open Korean LLM - Ko-Winogrande | 56.27 | Accuracy (%) | 25.3 |
| pfgen-bench - QA Mode - Fluency | 0.47 | Fluency Score | 24.8 |
| Open Korean LLM - Ko-EQ-Bench | 0 | EQ-Bench score | 24.5 |
| pfgen-bench - QA Mode - Score | 0.37 | pfgen Score (mean of three) | 23 |
| pfgen-bench - QA Mode - Helpfulness | 0.06 | Helpfulness Score | 22.4 |
Interactive version: theaggregate.ai/model?slug=xwin-lm-7b-v0-2 · How It Works · Data refreshed daily, snapshot 2026-09-19.