megamarcoroni-120B: benchmark results
An ibivibiv release whose card describes combining two Marcoroni 70B models. The published Llama configuration has 140 decoder layers; the release name alone does not establish learned parameter count. Provider: ibivibiv. Access: Open.
Aggregate Rating 1464 ± 37, rank #3101 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - HellaSwag | 88.94 | Normalized accuracy (%) (10-shot) | 95.2 |
| Open LLM Leaderboard v1 - MMLU | 69.88 | Accuracy (%) (5-shot) | 93 |
| Open LLM Leaderboard v1 - ARC Challenge | 72.01 | Normalized accuracy (%) (25-shot) | 90.7 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 64.24 | MC2 (%) (0-shot) | 82 |
| Open LLM Leaderboard v1 - WinoGrande | 80.9 | Accuracy (%) (5-shot) | 76.2 |
| Open LLM Leaderboard v1 - GSM8K | 21.53 | Accuracy (%) (5-shot) | 42 |
Interactive version: theaggregate.ai/model?slug=megamarcoroni-120b · How It Works · Data refreshed daily, snapshot 2026-11-08.