RandomMergeNoNormWEIGHTED-7B-DARETIES: benchmark results
jsfs11 release whose published recipe uses TIES merging with FelixChao/WestSeverus-7B-DPO-v2 as the base and CultriX/Wernicke-7B-v9, mlabonne/NeuralBeagle14-7B as additional inputs. Despite the release name, the embedded recipe specifies ties and normalize=true; no contrary method is inferred from the name. Provider: jsfs11. Access: Open.
Aggregate Rating 1466 ± 42, rank #3008 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - ARC Challenge | 73.38 | Normalized accuracy (%) (25-shot) | 97 |
| Open LLM Leaderboard v1 - HellaSwag | 88.5 | Normalized accuracy (%) (10-shot) | 92.4 |
| Open LLM Leaderboard v1 - GSM8K | 70.28 | Accuracy (%) (5-shot) | 92.3 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 71.5 | MC2 (%) (0-shot) | 91.9 |
| Open LLM Leaderboard v1 - WinoGrande | 83.58 | Accuracy (%) (5-shot) | 89.7 |
| Open LLM Leaderboard v1 - MMLU | 64.94 | Accuracy (%) (5-shot) | 78.9 |
Interactive version: theaggregate.ai/model?slug=randommergenonormweighted-7b-dareties · How It Works · Data refreshed daily, snapshot 2026-11-08.