> ## Documentation Index
> Fetch the complete documentation index at: https://docs.humainum.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Bring your own model (BYOK)

> Connect your own LLM provider and control what your builds run on.

Connect your own LLM provider credentials so builds run on infrastructure and terms you control, instead of a fixed provider Humainum chooses for you.

<Steps>
  <Step title="Open your project settings">
    From within a project, open the settings for AI/model credentials.
  </Step>

  <Step title="Add your provider's API key">
    Connect the credentials for the LLM provider you want your work to run on.
  </Step>

  <Step title="Build as usual">
    Everything else about the workflow stays the same — describe what you want, the agent plans and writes it — it's just running on your own connection now.
  </Step>
</Steps>

This is one of the reasons a Humainum project is never locked in: your provider, your terms, your real GitHub repository. See [Ownership, code & security](/ownership-and-security).