Captain_BMO-12B: benchmark results
Nitral-AI's experimental fine-tune of Mistral Nemo 12B Instruct on a GU_instruct data subset, released as a one-off internal test. Provider: Nitral-AI. Released 2024-11-01. Access: Open.
Aggregate Rating 1501 ± 46, rank #2342 of 10916 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard - GPQA | 31.96 | Score | 74.8 |
| Open LLM Leaderboard - BBH | 52.86 | Score | 63.5 |
| Open LLM Leaderboard - MATH Level 5 | 13.97 | Score | 60.1 |
| Open LLM Leaderboard - IFEval | 47.51 | Score | 54.9 |
| Open LLM Leaderboard - MMLU-Pro | 35.69 | Score | 53.6 |
| UGI - Writing | 34.35 | Writing Score | 51.5 |
| UGI - Willingness (W/10) | 4.5 | W/10 Score | 35.6 |
| UGI - Natural Intelligence | 18.38 | NatInt Score | 33.5 |
| Open LLM Leaderboard - MuSR | 37.48 | Score | 26.4 |
| UGI Leaderboard | 25.69 | UGI Score | 20.1 |
Interactive version: theaggregate.ai/model?slug=captain-bmo-12b · How It Works · Data refreshed daily, snapshot 2026-11-08.