Hermes 3, developed by Nous Research, is the latest large language model (LLM) in the Hermes series. Compared to Hermes 2, it shows significant improvements in agent capabilities, role-playing, reasoning, multi-turn dialogue, and long text coherence. The core philosophy of the Hermes series is to align the LLM with the user, providing end-users with strong guidance capabilities and control. Building on Hermes 2, Hermes 3 further enhances function calling and structured output capabilities, improving its general assistant capabilities and code generation skills.