Docs/API access
OpenAI compatibility
Point the base_url at APIONE and keep your existing SDK, request body and response handling.
On this page
APIONE's primary interface implements the OpenAI request/response format exactly. Any library or tool that lets you override base_url works without an adapter layer.
Configuration
| Setting | Value |
|---|---|
base_url / OPENAI_BASE_URL | https://api.apione.ai/v1 |
api_key / OPENAI_API_KEY | YOUR_APIONE_KEY |
python
from openai import OpenAI
client = OpenAI(
base_url="https://api.apione.ai/v1",
api_key="YOUR_APIONE_KEY",
)
resp = client.chat.completions.create(
model="gpt-5.6-sol",
messages=[{"role": "user", "content": "Write quicksort in Python"}],
temperature=0.2,
max_tokens=800,
)
print(resp.choices[0].message.content)Supported endpoints
| Endpoint | Notes |
|---|---|
POST /v1/chat/completions | Chat completion — the main endpoint, with streaming and tool calls |
GET /v1/models | Models callable with the current key |
POST /v1/responses | OpenAI Responses protocol |
POST /v1/images/generations | Image generation |
Differences from OpenAI
- Non-OpenAI models expose their own capabilities: sending
toolsto a model without function calling returns400. Capability matrices live on each model page in the model plaza. - Org/project fields are ignored:
organizationandprojecthave no effect — quota follows the APIONE key.
Tip
The cheapest migration is environment-only: set OPENAI_BASE_URL=https://api.apione.ai/v1 and OPENAI_API_KEY=YOUR_APIONE_KEY, and leave the code alone; take model IDs from the model list.
Related
Contact us
Something unclear, or need a human? Pick the channel that fits your question.
Supportsupport@apione.ai
WeCom
Telegram
Communityhttps://t.me/apione
Support tickets
Billing & accountConsole › Feedback