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

# Connect a model

> Use a Valkyrie-hosted model as the LLM inside your tools, agents, and SDKs.

This section is about **using a Valkyrie-hosted model for inference**, pointing a
client at a model you've [deployed](/guides/deploy-model) so it answers your prompts.

<Info>
  Want to **operate Valkyrie from an AI agent** instead, deploy models, stop them,
  read logs, manage jobs? That's a different integration: the **[MCP server](/mcp/overview)**.
  Connecting a model (this section) and connecting the MCP server (that section) are
  separate things and don't depend on each other.
</Info>

## Two compatible protocols

Valkyrie exposes your deployed model over two protocols, so most clients work
out of the box:

<CardGroup cols={2}>
  <Card title="OpenAI-compatible" icon="plug">
    The Chat Completions API. Works with the OpenAI SDKs and any OpenAI-compatible
    client, Cursor, LibreChat, editors, custom apps, and more.
  </Card>

  <Card title="Anthropic-compatible" icon="a">
    Valkyrie can also speak the Anthropic Messages protocol, so Claude Code can point
    at a Valkyrie-hosted model.
  </Card>
</CardGroup>

## What you'll need

* A **ready deployment** (or an [alias](/guides/aliases) pointing at one).
* A **deployment API key** (`vk_dep_`) with access to that model, see
  [API Keys](/dashboard/api-keys).
* Your **API base URL** (e.g. `https://valkyrie-back.azumo.com`).

## Pick your client

<CardGroup cols={3}>
  <Card title="Claude Code" icon="terminal" href="/connect/claude-code">
    Point your coding agent's model at a Valkyrie deployment.
  </Card>

  <Card title="Cursor" icon="code" href="/connect/cursor">
    Add a Valkyrie model as a custom OpenAI-compatible model.
  </Card>

  <Card title="OpenAI-compatible clients" icon="plug" href="/connect/openai-compatible">
    The generic recipe for any OpenAI-compatible SDK or app.
  </Card>
</CardGroup>

<Note>
  A self-hosted open model is not a frontier model. Behavior, tool-use quality, and
  context length depend entirely on the model you deployed, size the deployment for
  the workload you expect.
</Note>