Quickstart

1. Create an API key

1. Sign in to TAIBLES.
2. Open Settings → API Keys.
3. Create a key and copy it immediately — the full token is shown only once.

2. Call your default assistant

The first example targets an agent (default-assistant), not a raw LLM. The assistant runs with the tools, knowledge, and capabilities configured in TAIBLES.

Replace BASE_URL with the value from Settings (for example https://taibles.ai/api/openai/v1) and YOUR_API_KEY with your key.

cURL

curl "$BASE_URL/chat/completions" \
  -H "Authorization: Bearer $YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "default-assistant",
    "messages": [
      {"role": "user", "content": "What can you help me with?"}
    ]
  }'

Python

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://taibles.ai/api/openai/v1",  # or your host
)

completion = client.chat.completions.create(
    model="default-assistant",
    messages=[{"role": "user", "content": "What can you help me with?"}],
)
print(completion.choices[0].message.content)

JavaScript

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "YOUR_API_KEY",
  baseURL: "https://taibles.ai/api/openai/v1", // or your host
});

const completion = await client.chat.completions.create({
  model: "default-assistant",
  messages: [{ role: "user", content: "What can you help me with?" }],
});
console.log(completion.choices[0].message.content);

3. Optional: call a plain chat model

To use a specific LLM without selecting a custom agent, pass a chat model URL from GET /models. The request still runs through the default assistant with that model for the turn:

curl "$BASE_URL/chat/completions" \
  -H "Authorization: Bearer $YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "openai://gpt-5.4",
    "messages": [
      {"role": "user", "content": "Hello!"}
    ]
  }'

Exact model ids depend on what is configured for your deployment — list them with GET /models.

4. Continue reading

• Agents and models — custom agents by slug or UUID
• Chat completions — streaming, conversation continuity, builtin tools
• Examples — more SDK patterns