MoEv4Config-TestWeightedTIES-7B: benchmark results
jsfs11 release whose published recipe uses TIES merging with Kukedlc/NeuTrixOmniBe-7B-model-remix as the base and PetroGPT/WestSeverus-7B-DPO, vanillaOVO/supermario_v4 as additional inputs. Its configuration is a dense Mistral decoder, not a mixture-of-experts model despite the release name. Provider: jsfs11. Access: Open.
Aggregate Rating 1463 ± 40, rank #3139 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - GSM8K | 72.78 | Accuracy (%) (5-shot) | 98.1 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 70.87 | MC2 (%) (0-shot) | 91 |
| Open LLM Leaderboard v1 - WinoGrande | 83.82 | Accuracy (%) (5-shot) | 90.8 |
| Open LLM Leaderboard v1 - HellaSwag | 88.19 | Normalized accuracy (%) (10-shot) | 90.1 |
| Open LLM Leaderboard v1 - ARC Challenge | 71.59 | Normalized accuracy (%) (25-shot) | 89.6 |
| Open LLM Leaderboard v1 - MMLU | 65.07 | Accuracy (%) (5-shot) | 81.3 |
Interactive version: theaggregate.ai/model?slug=moev4config-testweightedties-7b · How It Works · Data refreshed daily, snapshot 2026-11-08.