Mahou-1.0-llama3-8B: benchmark results
A DPO fine-tune released by flammen.ai of nbeerbower/llama-3-stella-truthy-dpo-8B, using flammenai/FlameMix-DPO-v1. The release targets casual conversation and character roleplay. Provider: flammen.ai. Access: Open.
Aggregate Rating 1453 ± 34, rank #3636 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - GSM8K | 72.55 | Accuracy (%) (5-shot) | 97.9 |
| Open LLM Leaderboard v1 - MMLU | 68.48 | Accuracy (%) (5-shot) | 91.5 |
| Open LLM Leaderboard v1 - ARC Challenge | 69.45 | Normalized accuracy (%) (25-shot) | 83.2 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 59.05 | MC2 (%) (0-shot) | 73.1 |
| Open LLM Leaderboard v1 - HellaSwag | 84.95 | Normalized accuracy (%) (10-shot) | 70.4 |
| Open LLM Leaderboard v1 - WinoGrande | 78.06 | Accuracy (%) (5-shot) | 60.4 |
Interactive version: theaggregate.ai/model?slug=mahou-1-0-llama3-8b · How It Works · Data refreshed daily, snapshot 2026-11-08.