NousResearch
Hermes
Nous Research's community-driven fine-tune series. Hermes 3 sits on top of Llama 3.1 base weights, tuned for very strong tool use and steerable roleplay behaviour.
Visit homepage ↗History & context
NousResearch is a community-driven research collective known for high-quality fine-tunes on top of Meta's Llama base models. The Hermes line has been the flagship since Hermes 2 in 2023.
Hermes 3 (August 2024) sits on top of Llama 3.1 base weights across 8B, 70B and 405B sizes. The tuning recipe prioritises tool use and function calling; the resulting model is a common choice for agent-heavy workloads where reliable JSON output matters more than raw academic-benchmark quality.
Because Hermes builds on Llama 3.1, the effective licence is the Llama 3 Community Licence — the same 700M-MAU threshold applies.
Flagship model
Release timeline
2 models in this family
Larger Hermes 3 variant on top of Llama 3.1 70B. Widely used in agent-heavy workloads that need strong tool use combined with reliable function-calling schemas.
- Context
- 128K
- License
- llama-3
- VRAM Q4
- 42 GB
NousResearch's community-driven fine-tune on the Llama 3.1 8B base. Tuned for strong tool use, function calling and steerable persona behaviour. Inherits Llama 3's community licence and its 128K context.
- Context
- 128K
- License
- llama-3
- VRAM Q4
- 4.8 GB