openbuddy-zero-56B-v21.2-32k: benchmark results
OpenBuddy's 56B multilingual chatbot built from Yi-34B-32K and Llama 3 70B Instruct weights, further trained with 32k context. Provider: OpenBuddy. Released 2024-06-10. Access: Open.
Aggregate Rating 1569 ± 42, rank #1415 of 10916 rated models, from 14 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Chinese LLM Leaderboard | 67.64 | Official Nonzero-Task Mean (%) | 88.5 |
| Open Chinese LLM - C-Eval Semantic | 88.23 | Accuracy (%) | 87.9 |
| Open LLM Leaderboard - BBH | 61.28 | Score | 84.7 |
| Open Chinese LLM - HellaSwag | 67.78 | Accuracy (%) | 84.5 |
| Open LLM Leaderboard - MMLU-Pro | 43.99 | Score | 81.2 |
| Open Chinese LLM - ARC Challenge | 58.53 | Accuracy (%) | 80.8 |
| Open Chinese LLM - CMMLU | 68.34 | Accuracy (%) | 80.5 |
| Open Chinese LLM - WinoGrande | 66.85 | Accuracy (%) | 80.5 |
| Open Chinese LLM - GSM8K | 54.51 | Accuracy (%) | 74.7 |
| Open Chinese LLM - TruthfulQA MC | 58.07 | Accuracy (%) | 74.6 |
| Open LLM Leaderboard - GPQA | 31.8 | Score | 73.8 |
| Open LLM Leaderboard - MuSR | 43.05 | Score | 70.8 |
Interactive version: theaggregate.ai/model?slug=openbuddy-zero-56b-v21-2-32k · How It Works · Data refreshed daily, snapshot 2026-11-08.