Blur-7B-slerp-v0.1: benchmark results
liminerity's Blur-7B-slerp-v0.1 is a SLERP merge of OpenPipe/mistral-ft-optimized-1218, mlabonne/Marcoro14-7B-slerp. The released configuration uses 32 mistral decoder layers; merge recipe dtype is not a measured evaluator precision. Provider: liminerity. Access: Open.
Aggregate Rating 1446 ± 35, rank #3921 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - GSM8K | 72.1 | Accuracy (%) (5-shot) | 97.2 |
| Open LLM Leaderboard v1 - MMLU | 65.18 | Accuracy (%) (5-shot) | 82.7 |
| Open LLM Leaderboard v1 - HellaSwag | 86.58 | Normalized accuracy (%) (10-shot) | 81.9 |
| Open LLM Leaderboard v1 - ARC Challenge | 68.77 | Normalized accuracy (%) (25-shot) | 81.3 |
| Open LLM Leaderboard v1 - WinoGrande | 81.14 | Accuracy (%) (5-shot) | 77.4 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 60.64 | MC2 (%) (0-shot) | 76.3 |
Interactive version: theaggregate.ai/model?slug=blur-7b-slerp-v0-1 · How It Works · Data refreshed daily, snapshot 2026-11-08.