Hermes-2-Theta-Llama-3-8B [float16] [rev: dd2bfa013380]: benchmark results
Nous Research's Hermes-2-Theta-Llama-3-8B release evaluated at checkpoint revision dd2bfa013380639acf9e8fa45ceacbc45fb44081. This row preserves its recorded precision and chat-template setting; it is not merged with the current checkpoint or another evaluation mode. Provider: Nous Research. Access: Open.
Aggregate Rating 1427 ± 34, rank #4808 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - GSM8K | 70.43 | Accuracy (%) (5-shot) | 93 |
| Open LLM Leaderboard v1 - MMLU | 66.55 | Accuracy (%) (5-shot) | 87.5 |
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 55.66 | MC2 (%) (0-shot) | 65.7 |
| Open LLM Leaderboard v1 - ARC Challenge | 63.14 | Normalized accuracy (%) (25-shot) | 61.5 |
| Open LLM Leaderboard v1 - HellaSwag | 82.58 | Normalized accuracy (%) (10-shot) | 52.5 |
| Open LLM Leaderboard v1 - WinoGrande | 76.64 | Accuracy (%) (5-shot) | 49.6 |
Interactive version: theaggregate.ai/model?slug=hermes-2-theta-llama-3-8b-float16-rev-dd2bfa013380 · How It Works · Data refreshed daily, snapshot 2026-11-08.