Hermes 4.3 36B

A 36B hybrid reasoning model trained across a decentralized network. Built on ByteDance Seed-OSS-36B-Base.

Training and model behavior

Psyche coordinates decentralized training across the internet. Hermes 4.3 retains reasoning controls, function calling and structured output from Hermes 4.Hermes 4.3 36B model card

Capabilities

  • Hybrid reasoning

    Control reasoning behavior.

  • Function calling

    Connect tools and actions.

  • Structured output

    Produce structured responses.

Working with this model

Practical guidance. Use the decentralized release for application comparisons. Use the centralized control to study the training method.

Guidance. Different training runs should not be treated as interchangeable files.

Download formats

Nous also publishes GGUF packages for this model. These are deployment formats of the same release.Hermes 4.3 36B GGUF package

Specifications & access

Hermes 4.3 36B specifications
DetailValue and conditions
Launch date
AccessDownloadable
Model identifiersNousResearch/Hermes-4.3-36B
Context window524,288 tokens

Configured position limit; practical serving limits depend on the runtime.

Hermes 4.3 36B configuration
Repository license labelapache-2.0Hermes 4.3 36B model card

Related models

  • A Llama 3.1 405B derivative with optional reasoning and structured output.

    Launch date Downloadable
  • A Qwen3-14B derivative with optional reasoning and structured output.

    Launch date September 2025Downloadable
  • A Llama 3.1 70B derivative with optional reasoning and structured output.

    Launch date Downloadable