openbuddy-qwen2.5llamaify-14B-v23.1-200k: benchmark results
OpenBuddy's multilingual assistant fine-tune of Qwen2.5 14B converted to Llama architecture, with a 200K-token context window. Provider: OpenBuddy. Released 2024-09-23. Access: Open.
Aggregate Rating 1458 ± 19, rank #952 of 1661 rated models, from 21 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Korean LLM - Ko-GSM8K (flexible extract) | 59.74 | Exact match, flexible extract (%) | 93.5 |
| Open Korean LLM Leaderboard | 43.62 | Average Score (%) | 88.4 |
| Open Ko-LLM Leaderboard | 43.62 | Average (%) | 88.3 |
| Open Korean LLM - Ko-IFEval | 43.25 | Strict accuracy, prompt/instruction mean (%) | 87.6 |
| Open LLM Leaderboard - MMLU-Pro | 46.73 | Score | 85.7 |
| Open Korean LLM - KorNAT-Helpful | 50.03 | Normalized accuracy (%) | 83.7 |
| Open Portuguese LLM - BLUEX | 63.42 | Accuracy (%) | 83.5 |
| Open Portuguese LLM - OAB Exams | 53.35 | Accuracy (%) | 83.3 |
| Open LLM Leaderboard - BBH | 60.13 | Score | 82 |
| Open LLM Leaderboard - GPQA | 33.31 | Score | 81.7 |
| Open Portuguese LLM - ENEM | 71.66 | Accuracy (%) | 81.1 |
| Open Korean LLM - KorNAT-Harmless | 65.41 | Normalized accuracy (%) | 80 |
Interactive version: theaggregate.ai/model?slug=openbuddy-qwen2-5llamaify-14b-v23-1-200k · How It Works · Data refreshed daily, snapshot 2026-10-23.