MN-LooseCannon-12B-v1: benchmark results
GalrionSoftworks' TIES merge of Starcannon v3 and Lyra v1, two roleplay-oriented Mistral Nemo 12B tunes. Provider: Other. Released 2024-08-09. Access: Open.
Unified ELO 1532 ± 1, rank #535 of 1392 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard - IFEval | 54.18 | Score | 65 |
| UGI - Writing | 37.49 | Writing Score | 63.6 |
| Open LLM Leaderboard - MuSR | 10.96 | Score | 56.1 |
| UGI - Natural Intelligence | 22.82 | NatInt Score | 53.6 |
| Open LLM Leaderboard - BBH | 29.98 | Score | 51.7 |
| Open LLM Leaderboard - MMLU-Pro | 24.4 | Score | 43.6 |
| Open LLM Leaderboard - MATH Level 5 | 8.53 | Score | 42.8 |
| Open LLM Leaderboard - GPQA | 4.7 | Score | 40.2 |
| UGI - Willingness (W/10) | 4.8 | W/10 Score | 36.8 |
| UGI Leaderboard | 31.13 | UGI Score | 35.1 |
Interactive version: theaggregate.ai/model?slug=mn-loosecannon-12b-v1 · How It Works · Data refreshed daily, snapshot 2026-09-05.