LLM Reference

Hermes 2 Pro Mistral 7B

About

Hermes 2 Pro Mistral 7B is a cutting-edge large language AI model developed by Nous Research, based on the Mistral 7B architecture and comprising 7.24 billion parameters. It builds upon the previous Nous Hermes 2 model, offering enhanced data quality and new functionalities. Notable for its proficiency in conversation and task completion, Hermes 2 Pro excels in function calling with a remarkable 90-91% accuracy. It is also adept at generating structured JSON outputs with 84% accuracy and supports ChatML for structured multi-turn interactions. Trained on an enriched OpenHermes 2.5 Dataset, the model includes a special system prompt and a new ChatML role, ensuring reliability and compatibility with OpenAI APIs. Quantized GGUF format availability allows for versatile and efficient deployment across various hardware platforms.

Capabilities

MultimodalFunction CallingTool UseJSON Mode

Providers(2)

ProviderInput (per 1M)Output (per 1M)Type
Fireworks AI Platform
Provisioned
Cloudflare Workers AI
Serverless

Specifications

FamilyHermes 2
Parameters7B
ArchitectureDecoder Only
Specializationgeneral