DeepHermes 3 - Llama-3.1 8B Preview (Non-reasoning)
Unknown SizeBy Nous Research • Released 2025-02-13
Capability Radar
Avg Score
16
Across all benchmarks
Participated
7
Benchmarks
Benchmark Performance
| Benchmark | Category | Score |
|---|---|---|
| MMLU-Pro | Knowledge | 36.5 |
| GPQA Diamond | Knowledge | 27 |
| MATH-500 | Reasoning | 21.8 |
| SciCode | Reasoning Knowledge | 9.1 |
| LiveCodeBench | Coding | 8.5 |
| Artificial Analysis Intelligence Index | Knowledge | 7.6 |
| HLE | Knowledge Multi-Modal | 4.3 |