Skip to main content
Run Claude Code, or an Anthropic SDK, on your agent’s budget. Claude Code and an Anthropic SDK’s message-creation calls use POST /v1/messages, which speaks Anthropic’s Messages API.
  • The base URL is https://inference.agentic.anchorage.com, without /v1. The client adds /v1/messages itself.
  • The API key is the agent’s API key. Send it the way the client normally sends an Anthropic key.
  • model is a Claude model’s identifier from the model list.
The agent’s budget must admit the inference provider, and the agent needs an API key. See Before you start.

​
Claude Code

Set these environment variables before you start it:
Also set ANTHROPIC_MODEL and ANTHROPIC_DEFAULT_HAIKU_MODEL to Claude models from the list, so every call Claude Code makes names a model the list prices.

​
An Anthropic SDK

Pass the same two values as the client’s base_url (or baseURL) and api_key (or apiKey). The base URL has no /v1.

​
Long sessions

  • A streamed response is cut off after 120 seconds. Set CLAUDE_CODE_MAX_OUTPUT_TOKENS to keep long answers under that.
  • A request body is limited to 1.5 MiB. Run /compact to shrink a long session.
See Anthropic Messages API for the headers it forwards and the error shape it returns.