StarDust-12B-v2: benchmark results
Luni's roleplay-focused DARE TIES merge of Mistral Nemo 12B tunes, built on Lyra v3 with Magnum v2, Pantheon-RP, and bophades. Provider: Luni. Released 2024-09-01. Access: Open.
Aggregate Rating 1495 ± 46, rank #2436 of 10916 rated models, from 10 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard - MuSR | 43.38 | Score | 73.5 |
| Open LLM Leaderboard - BBH | 54.19 | Score | 69.8 |
| Open LLM Leaderboard - IFEval | 56.29 | Score | 68.2 |
| UGI - Willingness (W/10) | 5.8 | W/10 Score | 50.8 |
| Open LLM Leaderboard - MMLU-Pro | 34.39 | Score | 49.6 |
| Open LLM Leaderboard - GPQA | 29.36 | Score | 48.7 |
| UGI - Writing | 32.79 | Writing Score | 46.4 |
| Open LLM Leaderboard - MATH Level 5 | 6.87 | Score | 36.9 |
| UGI Leaderboard | 30.51 | UGI Score | 32.3 |
| UGI - Natural Intelligence | 17.4 | NatInt Score | 29.7 |
Interactive version: theaggregate.ai/model?slug=stardust-12b-v2 · How It Works · Data refreshed daily, snapshot 2026-11-08.