Phigments12: benchmark results
liminerity's Phigments12 is a SLERP merge of liminerity/merge6, liminerity/merge3. The released configuration uses 32 phi decoder layers; merge recipe dtype is not a measured evaluator precision. Provider: liminerity. Access: Open.
Aggregate Rating 1384 ± 35, rank #6734 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - GSM8K | 61.33 | Accuracy (%) (5-shot) | 73.9 |
| Open LLM Leaderboard v1 - ARC Challenge | 62.63 | Normalized accuracy (%) (25-shot) | 59 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 51.71 | MC2 (%) (0-shot) | 53 |
| Open LLM Leaderboard v1 - MMLU | 58.43 | Accuracy (%) (5-shot) | 43.3 |
| Open LLM Leaderboard v1 - WinoGrande | 74.66 | Accuracy (%) (5-shot) | 37.9 |
| Open LLM Leaderboard v1 - HellaSwag | 77.1 | Normalized accuracy (%) (10-shot) | 30.2 |
Interactive version: theaggregate.ai/model?slug=phigments12 · How It Works · Data refreshed daily, snapshot 2026-11-08.