Hermes 4 is a large-scale reasoning model built on Meta-Llama-3.1-405B and released by Nous Research. It introduces a hybrid reasoning mode, where the model can choose to deliberate internally with...
Specifications
- Provider
- Nous
- Type
- Open-source / open-weight
- Modality
- text->text
- Context window
- 131,072
- Knowledge cutoff
- 2024-08-31
- Released
- August 26, 2025
Capabilities
Input: textOutput: text
Strengths
frequency_penaltyinclude_reasoningmax_tokenspresence_penaltyreasoningrepetition_penaltyresponse_formattemperaturetop_ktop_p
