openbuddy-mistral-22B-v21.1-32k: benchmark results
OpenBuddy's multilingual chatbot fine-tune of Vezora's Mistral-22B-v0.2, extended to a 32K context as part of the open-source OpenBuddy chat model family. Provider: OpenBuddy. Released 2024-05-17. Access: Open.
Aggregate Rating 1295 ± 17, rank #1493 of 1656 rated models, from 24 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - GSM8K | 68.84 | Accuracy (%) (5-shot) | 87.1 |
| Open Korean LLM - Ko-GPQA-Diamond | 26.26 | Normalized accuracy (%) | 85.5 |
| Open LLM Leaderboard v1 - MMLU | 64.77 | Accuracy (%) (5-shot) | 76.3 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 55.31 | MC2 (%) (0-shot) | 64.8 |
| Open Korean LLM - KorNAT-CKA | 26.17 | Normalized accuracy (%) | 62.5 |
| Open LLM Leaderboard v1 - ARC Challenge | 60.67 | Normalized accuracy (%) (25-shot) | 48.9 |
| Open LLM Leaderboard v1 - WinoGrande | 75.69 | Accuracy (%) (5-shot) | 43.5 |
| Open Chinese LLM - TruthfulQA MC | 51.24 | Accuracy (%) | 26.3 |
| Open Korean LLM - Ko-EQ-Bench | 0 | EQ-Bench score | 24.5 |
| Open LLM Leaderboard v1 - HellaSwag | 67.81 | Normalized accuracy (%) (10-shot) | 18.8 |
| Open Korean LLM - Ko-IFEval | 19.66 | Strict accuracy, prompt/instruction mean (%) | 16.7 |
| Open Korean LLM Leaderboard | 28.96 | Average Score (%) | 8.7 |
Interactive version: theaggregate.ai/model?slug=openbuddy-mistral-22b-v21-1-32k · How It Works · Data refreshed daily, snapshot 2026-10-20.