Llama-3.2-Rabbit-Ko-3B-Instruct: benchmark results
CarrotAI's Korean-and-English instruction fine-tune of Meta's Llama 3.2 3B Instruct, SFT-trained on Korean instruction data. Provider: Meta. Released 2024-09-30. Access: Open.
Unified ELO 1479 ± 1, rank #895 of 1553 rated models, from 7 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard - IFEval | 71.99 | Score | 85.6 |
| Open Korean LLM Leaderboard | 40.24 | Average Score (%) | 73.8 |
| Open LLM Leaderboard - MATH Level 5 | 20.54 | Score | 72 |
| Open LLM Leaderboard - BBH | 21.5 | Score | 32.1 |
| Open LLM Leaderboard - MMLU-Pro | 20.25 | Score | 31.6 |
| Open LLM Leaderboard - GPQA | 2.8 | Score | 25.8 |
| Open LLM Leaderboard - MuSR | 3.98 | Score | 19.6 |
Interactive version: theaggregate.ai/model?slug=llama-3-2-rabbit-ko-3b-instruct · How It Works · Data refreshed daily, snapshot 2026-09-08.