OpenAI Compatible API Gateway
OpenAI-Compatible API Gateway: Change the Base URL and Go
Most AI apps, SDKs and clients let you set a custom OpenAI base URL. Set it to https://gw-link.com/v1, use your gw-link sk- key, and existing code keeps working. The same gateway also serves the native Anthropic and Gemini APIs.
Full OpenAI surface
/v1/chat/completions, /v1/responses, /v1/embeddings, /v1/images/generations and /v1/models — chat, agents, embeddings and images.
Several protocols, one host
The same domain also serves Anthropic /v1/messages and Gemini /v1beta natively, so each tool uses what it expects.
Central control
Per-key spending caps, per-group model access, and usage and cost in one console.
Setup
- Create an API key in the gw-link console.
- Set your client's or SDK's base URL to https://gw-link.com/v1.
- Use your sk- key as the API key and pick a model ID from /v1/models.
- Send a test request to confirm it works.
Configuration examples
curl
curl https://gw-link.com/v1/chat/completions \
-H "Authorization: Bearer sk-your-gw-link-key" \
-H "Content-Type: application/json" \
-d '{"model":"gpt-5.5","messages":[{"role":"user","content":"hello"}]}'Python (openai SDK)
from openai import OpenAI
client = OpenAI(base_url="https://gw-link.com/v1", api_key="sk-your-gw-link-key")
resp = client.chat.completions.create(
model="gpt-5.5",
messages=[{"role": "user", "content": "hello"}],
)
print(resp.choices[0].message.content)FAQ
What does OpenAI-compatible mean?
The paths, request bodies and responses match the OpenAI API, so any OpenAI client with a custom base URL works after you swap base_url and API key.
Do Cherry Studio, Chatbox, Cline or Continue work?
Yes. Choose the OpenAI or OpenAI-compatible provider, set the base URL to https://gw-link.com/v1 (some clients want just https://gw-link.com) and paste your gw-link key.
Which auth header should I use?
Standard Authorization: Bearer sk-... The gateway also accepts x-api-key and x-goog-api-key.
How do I see available models?
Call GET https://gw-link.com/v1/models — it returns the model IDs your key can use.
Last updated: