API Integration

The API matches the OpenAI chat completions protocol. Request bodies are forwarded upstream verbatim; no fields are stripped.

Base URL：https://aiapigpt.com/v1
Authentication: API key (starts with sk-)
Protocol: OpenAI chat completions (/v1/chat/completions)

Replace sk-your-sub2api-key with your own key. Keep it safe and never commit it to a repository.

1. Base URL and headers

Send Authorization: Bearer <your API key> with every request. An invalid key and a disabled key return different status codes (401 / 403), and neither reveals whether the key exists.

Base URL

Authorization: Bearer sk-your-sub2api-key
Content-Type: application/json

2. Endpoints

Endpoints	Notes
POST /v1/chat/completions	Chat completion. With stream: true the response is streamed as SSE chunks.
GET /v1/models	List the models you can call right now (authentication required). Only models that are priced, backed by a usable upstream account, and within your key's allowlist are returned — so nothing in the list is un-callable.
GET /healthz	Health check. No authentication required.
Any other path	Always 404. No other OpenAI endpoints are implemented (images, audio, embeddings, etc.).

3. Example request

3. 1 curl

curl https://aiapigpt.com/v1/chat/completions \
  -H "Authorization: Bearer sk-your-sub2api-key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4o",
    "messages": [{"role": "user", "content": "Hello"}]
  }'

3. 2 Python SDK (openai)

from openai import OpenAI

client = OpenAI(
    base_url="https://aiapigpt.com/v1",
    api_key="sk-your-sub2api-key",
)

resp = client.chat.completions.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": "Hello"}],
)
print(resp.choices[0].message.content)

4. Useful response headers

X-Request-Id: The request ID. You may also send your own X-Request-Id header to line up with your own logs.
X-Sub2Api-Charge: ok = charged; pending = charging failed and was queued for replay, which retries automatically.