Skip to main content
Point an OpenAI-compatible client at Anchorage Agentic Banking.

​
Before you start

  • An agent whose budget admits the inference provider. The default allow-all budget does.
  • An API key for that agent. See Mint a key.

​
Pick a model

Browse the Inference page, or list models from GET /v1/models with no key. See How the Inference page shows prices for what each entry carries.

​
Set up the client

The Inference page’s How to connect card walks through these three values with ready-to-paste snippets for curl, Python, and TypeScript. Under the card, Connect your harness shows the block to add the endpoint as a custom provider in Hermes or in OpenClaw. The Agents page and each agent’s page link there with Set up inference.

​
Control repeat requests (optional)

Send an Idempotency-Key header to control whether a repeated request shares a charge or pays again. See Managing idempotency keys.

​
Raise your client’s timeout

Give your client more than three minutes, so it doesn’t give up on a call you already paid for. To generate an image or a video instead, see Generate an image or video. For speech, sound effects, or music, see Generate speech, sound effects, or music. To run Claude Code or an Anthropic SDK, see Use Claude Code or an Anthropic SDK. For a free typed judgment, see Ask for a typed judgment. See Paid inference for the full flow, and Errors and retries for the failure modes, including over-budget declines.