Access Endpoint

Unified Service Endpoint

CoAIWork exposes a single service endpoint:

https://coaiwork.com

Client-specific base URL differences are protocol-specific adaptations of the same endpoint.

Client Mapping

The Request path column is relative to the base URL configured in the client. The Full request URL example shows the final URL sent to CoAIWork.

Client	Value to configure	Protocol	Request path relative to that value	Full request URL example
Claude Code	https://coaiwork.com	Anthropic Messages	/v1/messages	https://coaiwork.com/v1/messages
Codex	https://coaiwork.com/v1	OpenAI Responses	/responses	https://coaiwork.com/v1/responses
OpenCode	https://coaiwork.com/v1	OpenAI Chat Completions	/chat/completions	https://coaiwork.com/v1/chat/completions
Qwen Code	https://coaiwork.com/v1	OpenAI Compatible	/chat/completions	https://coaiwork.com/v1/chat/completions

Authentication

OpenAI-compatible

• Header: Authorization: Bearer <TOKEN>
• If the client base URL is https://coaiwork.com/v1, use:
• GET /models
• POST /chat/completions
• POST /responses

Anthropic-compatible

• Header: x-api-key: <TOKEN>
• Header: anthropic-version: 2023-06-01
• If the client base URL is https://coaiwork.com, use:
• GET /v1/models
• POST /v1/messages

Server Absolute Paths

The table below lists the server's full paths relative to the domain root https://coaiwork.com.

Path	Method	Purpose
/v1/models	GET	Model visibility check
/v1/chat/completions	POST	OpenAI Chat Completions
/v1/responses	POST	OpenAI Responses
/v1/messages	POST	Anthropic Messages
/api/usage/token	GET	Token usage and model limits

Configuration Rule

• Use https://coaiwork.com when a client requires a service endpoint or Anthropic-style base URL
• Use https://coaiwork.com/v1 when a client requires an OpenAI-compatible base URL
• When the base URL is already https://coaiwork.com/v1, the client should request /models, /chat/completions, or /responses relative to that base URL. Do not add another /v1.
• Use model names exactly as returned by GET /v1/models