Mistral-Nemo-Instruct-2407-abliterated: benchmark results
Uncensored community variant of Mistral NeMo Instruct 2407 with refusal directions removed via abliteration. Provider: natong19. Released 2024-08-15. Access: Open.
Aggregate Rating 1521 ± 19, rank #713 of 1656 rated models, from 25 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Ko-LLM Leaderboard | 48.85 | Average (%) | 96.1 |
| Open Korean LLM Leaderboard | 48.85 | Average Score (%) | 96.1 |
| Open Korean LLM - Ko-Winogrande | 69.93 | Accuracy (%) | 95.9 |
| Open Korean LLM - Ko-GSM8K (flexible extract) | 62.32 | Exact match, flexible extract (%) | 95.4 |
| Open Korean LLM - Ko-EQ-Bench | 50.9 | EQ-Bench score | 95.2 |
| Open Korean LLM - Ko-IFEval | 52.3 | Strict accuracy, prompt/instruction mean (%) | 95.2 |
| Open Portuguese LLM - FaQuAD NLI | 82.47 | Macro F1 (%) | 94.6 |
| Open Korean LLM - Ko-GPQA-Diamond | 26.77 | Normalized accuracy (%) | 88.9 |
| Open Korean LLM - KorNAT-Helpful | 50.43 | Normalized accuracy (%) | 85.6 |
| Open Portuguese LLM - OAB Exams | 54.17 | Accuracy (%) | 85.3 |
| Open Portuguese LLM - ASSIN2 RTE | 92.85 | Macro F1 (%) | 84.4 |
| Open Portuguese LLM - ENEM | 71.94 | Accuracy (%) | 81.7 |
Interactive version: theaggregate.ai/model?slug=mistral-nemo-instruct-2407-abliterated · How It Works · Data refreshed daily, snapshot 2026-10-20.