tools and tool_calls fields. Holo-specific behavior (reasoning, structured outputs, and the coordinate convention) comes through a few extra fields documented here.
Endpoints
POST /chat/completions
The inference endpoint: parameters, response fields, streaming.
GET /v1/models
Discover served models, limits, pricing, and deprecation dates at runtime.
Endpoint and auth
The Quickstart shows the client setup in Python, TypeScript, and cURL. Model IDs, per-model limits, pricing, and tiers live on the Models page.
The two response channels
Every call returnsmessage.content (the action or answer) and message.reasoning (the thinking trace, read-only). Field-by-field details are in the chat completions reference; how to carry state across turns is in Core concepts.
Conventions
Coordinates in[0, 1000], the 3-screenshot image budget, choosing between structured outputs and function calling, and passing Holo-specific fields through the OpenAI SDKs are covered in Core concepts.
Rate limits
The credit dashboard shows your balance and usage. The balance is checked before a request runs, so a large request on a low balance can overshoot by a few cents; contact support@hcompany.ai if it keeps growing. Unused credits may expire one year after purchase and are not refundable by default, subject to regional obligations and the terms of use.
Errors
Next steps
Chat completions
Full parameter and response reference.
Models
IDs, limits, pricing, lifecycle.
Build an agent
The loop conventions Holo is trained on.