xteapot/mar-llama3.1-8B

TEXT GENERATIONPricing:Input $0.2 / Cached $0.028 / Output $0.32Concurrent Unit Cost:1Model Size:8BQuant:FP8Context Size:32kTool Calling:SupportedPublished:Sep 25, 2026License:llama3Architecture:Transformer Featherless Exclusive Cold

The xteapot/mar-llama3.1-8B model, also known as Hermes 3, is an 8 billion parameter generalist language model developed by Nous Research, based on the Llama-3.1 architecture. It features advanced agentic capabilities, improved roleplaying, reasoning, multi-turn conversation, and long context coherence, with a context length of 32768 tokens. This model is specifically aligned for user control and excels in function calling, structured output, and code generation.

Loading preview...

Hermes 3 - Llama-3.1 8B Overview

Hermes 3 is the latest iteration in Nous Research's Hermes series, built on the Llama-3.1 architecture. This 8 billion parameter model is designed as a generalist language model with significant enhancements over its predecessor, Hermes 2. A core philosophy of the Hermes series is to provide powerful steering capabilities and user control, aligning the LLM to specific user needs.

Key Capabilities

  • Advanced Agentic Capabilities: Improved ability to act as an intelligent agent.
  • Enhanced Roleplaying & Reasoning: Offers much better performance in roleplaying scenarios and complex reasoning tasks.
  • Multi-turn Conversation & Long Context Coherence: Excels in maintaining coherent and relevant dialogue over extended multi-turn conversations, supporting a 32768-token context length.
  • Reliable Function Calling & Structured Output: Features more robust and dependable function calling and structured output generation, compatible with OpenAI API formats.
  • Improved Code Generation: Demonstrates enhanced skills in generating code.
  • ChatML Prompt Format: Utilizes ChatML for structured multi-turn dialogue, including system prompts for steerability.

Good For

  • Agentic Applications: Developing applications requiring advanced agentic behavior.
  • Complex Conversational AI: Building chatbots or assistants that need to handle intricate multi-turn dialogues and maintain long-term coherence.
  • Function Calling & Tool Use: Integrating with external tools and APIs through reliable function calling.
  • Structured Data Generation: Generating JSON or other structured outputs based on defined schemas.
  • Code Generation Tasks: Assisting with various code generation requirements.

Hermes 3 is competitive with Llama-3.1 Instruct models in general capabilities, offering distinct strengths. For more in-depth information, refer to the Hermes 3 Technical Report.

Finetunes
1 models