patricide-12B-Unslop-Mell: benchmark results
redrix's SLERP merge of TheDrummer's UnslopNemo v4.1 and inflatebot's Mag-Mell R1, two Mistral Nemo 12B roleplay tunes. Provider: redrix. Released 2024-12-01. Access: Open.
Aggregate Rating 1532 ± 47, rank #1929 of 10916 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard - GPQA | 32.38 | Score | 77.5 |
| Open LLM Leaderboard - BBH | 53.99 | Score | 68.9 |
| UGI - Writing | 38.81 | Writing Score | 67.3 |
| Open LLM Leaderboard - MATH Level 5 | 13.14 | Score | 58 |
| UGI - Natural Intelligence | 23.59 | NatInt Score | 55.3 |
| Open LLM Leaderboard - MMLU-Pro | 35.7 | Score | 53.7 |
| Open LLM Leaderboard - MuSR | 40.26 | Score | 44.2 |
| UGI - Willingness (W/10) | 5.2 | W/10 Score | 43.1 |
| Open LLM Leaderboard - IFEval | 40.74 | Score | 40.9 |
| UGI Leaderboard | 28.84 | UGI Score | 27.6 |
Interactive version: theaggregate.ai/model?slug=patricide-12b-unslop-mell · How It Works · Data refreshed daily, snapshot 2026-11-08.