Hermes 3 405B
Open Weights
Nous Research
Nous Research's flagship open model (August 2024), the first full-parameter fine-tune of Llama 3.1 405B. Hermes 3 is a steerable generalist with strong agentic capabilities, roleplay, reasoning, and long-context coherence, competitive with or exceeding Llama 3.1 Instruct.
Strengths
- First full-parameter fine-tune of Llama 3.1 405B
- Highly steerable generalist
- Strong agentic capabilities and reasoning
- Excellent roleplay and long-context coherence
- Competitive with or exceeding Llama 3.1 Instruct
Caveats
- 405B is very demanding to self-host
- Text-only (no vision input)
- Community-driven support
- Vendor
- Nous Research
- Context Window
- 131,072 tokens
- Max Output
- 4,096 tokens
- License
- Llama 3.1 Community License
- Release Date
- 2024-08-16
- Modalities
- text
Capabilities
Vision
Audio
Video
Tool Use
Resources
Pricing
Open weights under the Llama 3.1 Community License - free to self-host; 405B is demanding to run. Available via inference providers.