Hermes 2 Pro - Llama-3 70B Overview

Hermes 2 Pro is a 70 billion parameter model, a collaborative effort by Nous Research, @interstellarninja, and Fireworks.AI. It builds upon Nous Hermes 2 with an enhanced dataset and a newly developed Function Calling and JSON Mode dataset. The model maintains strong general task and conversational abilities while significantly improving its performance in structured outputs.

Key Capabilities

• Function Calling: Achieves 90% on internal function calling evaluations, utilizing a specialized system prompt and multi-turn structure with new ChatML roles and dedicated tokens (<tools>, <tool_call>, <tool_response>).
• JSON Structured Outputs: Scores 84% on structured JSON output evaluations, designed to respond with only a JSON object based on a provided schema.
• General Task & Conversation: Retains excellent performance in broad language understanding and generation tasks.
• Agentic Capabilities: Incorporates single tokens for agentic parsing during streaming, enhancing reliability.

Good For

• Developers requiring robust function calling in their applications.
• Use cases demanding reliable JSON structured outputs from an LLM.
• General conversational AI and complex task execution where precise output formatting is critical.
• Applications leveraging agentic workflows that benefit from specialized parsing tokens.