OpenAI-compatible hosted API
Base URL: https://api.botconnector.id/v1
"OpenAI-compatible" is a claim that should be checkable, so here is exactly how far it goes. Two hosted routes are in the public contract:
GET /v1/models— the model IDs this account can callPOST /v1/chat/completions— chat completions, streamed or not
Everything else that an OpenAI SDK can call is outside the contract, and a request to it returns an explicit error rather than an answer from some other provider.
Pointing a client at it
Any OpenAI-compatible client works once the base URL and the key change:
from openai import OpenAI
import os
client = OpenAI(
base_url="https://api.botconnector.id/v1",
api_key=os.environ["BOTCONNECTOR_API_KEY"],
)import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://api.botconnector.id/v1",
apiKey: process.env.BOTCONNECTOR_API_KEY,
});Tools that take a base URL as configuration rather than in code need the same value: https://api.botconnector.id/v1, with the /v1 included.
Model IDs
Use a canonical ID from GET /v1/models. A display name from the catalogue is not an ID, and an ID that your account cannot reach is not callable — see GET /v1/models.
Beyond these two routes
The Anthropic Messages API and the native OpenAI Responses API are served on their own paths, with their own terms; the Hosted Developer API page lists them. OpenAI Files, Vector Stores, Embeddings, Rerank, Audio/Realtime, fine-tuning, Batch and administration resources are not available.