Hermes 3 70B
Hermes 3 70B is Nous Research's flagship instruction-following model, fine-tuned for advanced reasoning, creative writing, and complex task completion. It features exceptional instruction adherence and strong performance across multiple domains.
Benchmark results
| Benchmark | Score | Tags | Source |
|---|---|---|---|
| AGIEval | 56.2% | self-reported llm-stats | link → |
| ARC-C | 65.5% | self-reported llm-stats | link → |
| ARC-E | 83.0% | self-reported llm-stats | link → |
| BBH | 67.8% | self-reported llm-stats | link → |
| BoolQ | 88.0% | self-reported llm-stats | link → |
| GPQA | 66.1% | self-reported llm-stats | link → |
| HellaSwag | 88.2% | self-reported llm-stats | link → |
| IFBench | 81.2% | self-reported llm-stats | link → |
| MATH | 20.8% | self-reported llm-stats | link → |
| MMLU | 79.1% | self-reported llm-stats | link → |
| MMLU-Pro | 47.2% | self-reported llm-stats | link → |
| MT-Bench | 8.99 | self-reported llm-stats | link → |
| MuSR | 50.7% | self-reported llm-stats | link → |
| OpenBookQA | 49.4% | self-reported llm-stats | link → |
| PIQA | 84.4% | self-reported llm-stats | link → |
| TruthfulQA | 63.3% | self-reported llm-stats | link → |
| Winogrande | 83.2% | self-reported llm-stats | link → |