eleusis-7B-alpha: benchmark results
maldv Eleusis 7B Alpha is a nine-partition merge of OpenHermes 2.5 Mistral and other Hermes-related checkpoints, using different input/output SLERP gradients. It uses the ChatML prompt format. Provider: maldv. Access: Open.
Aggregate Rating 1424 ± 33, rank #4991 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - GSM8K | 61.79 | Accuracy (%) (5-shot) | 74.6 |
| Open LLM Leaderboard v1 - HellaSwag | 84.87 | Normalized accuracy (%) (10-shot) | 69.6 |
| Open LLM Leaderboard v1 - ARC Challenge | 64.93 | Normalized accuracy (%) (25-shot) | 68.5 |
| Open LLM Leaderboard v1 - WinoGrande | 79.16 | Accuracy (%) (5-shot) | 67.7 |
| Open LLM Leaderboard v1 - MMLU | 64.1 | Accuracy (%) (5-shot) | 65.7 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 54.34 | MC2 (%) (0-shot) | 62.2 |
Interactive version: theaggregate.ai/model?slug=eleusis-7b-alpha · How It Works · Data refreshed daily, snapshot 2026-11-08.