A single, OpenAI-compatible endpoint for every LucidQuery model. If you already use the OpenAI SDK, point the base URL at us and your existing code keeps working — streaming, tool calls, and JSON mode included.

Authenticate every request with a Bearer token in the Authorization header. Keys look like sk_lucidquery_… and are created in your dashboard. Keep them server-side — never ship a key in client code.

curl https://api.lucidquery.com/v1/chat/completions \
  -H "Authorization: Bearer $LUCIDQUERY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "model": "lucidquery-agi-01-swift", "messages": [{ "role": "user", "content": "Hi" }] }'

A missing or invalid key returns 401 with an authentication_error.

Install the OpenAI SDK, set the base URL and your key, then stream a completion.

from openai import OpenAI

client = OpenAI(
    api_key="lq_live_...",
    base_url="https://api.lucidquery.com/v1",
)

stream = client.chat.completions.create(
    model="lucidquery-agi-01-frontier",
    messages=[{"role": "user", "content": "Explain quantum error correction."}],
    stream=True,
)

for chunk in stream:
    print(chunk.choices[0].delta.content or "", end="")

The AGI 01 family covers general and flagship intelligence. Switch models with a single string; pricing is per million tokens, billed from your prepaid balance.

Additional specialized models are also reachable via GET /v1/models. The AGI 01 family above is recommended for general use.

Generate a model response for a conversation. Returns a chat.completion object, or a stream of chat.completion.chunk events when stream is true.

{
  "id": "chatcmpl_8f0c2a…",
  "object": "chat.completion",
  "created": 1781214488,
  "model": "lucidquery-agi-01-frontier",
  "choices": [
    {
      "index": 0,
      "message": { "role": "assistant", "content": "Hello!" },
      "finish_reason": "stop"
    }
  ],
  "usage": { "prompt_tokens": 15, "completion_tokens": 3, "total_tokens": 18 }
}

Set response_format: { "type": "json_object" } to guarantee the model returns a single valid JSON object. Describe the shape you want in the prompt; for best results include an example of the target JSON.

from openai import OpenAI

client = OpenAI(
    api_key="sk_lucidquery_...",
    base_url="https://api.lucidquery.com/v1",
)

resp = client.chat.completions.create(
    model="lucidquery-agi-01-swift",
    messages=[
        {"role": "system", "content": "Extract the city and country as JSON."},
        {"role": "user", "content": "I live in Lyon, France."},
    ],
    response_format={"type": "json_object"},
)

print(resp.choices[0].message.content)
# {"city": "Lyon", "country": "France"}

Supported types: json_object and text.

With stream: true, tokens arrive as server-sent events. Each line is a chat.completion.chunk with a delta; the stream terminates with data: [DONE].

data: {"id":"chatcmpl_8f0c2a…","object":"chat.completion.chunk","created":1781214488,"model":"lucidquery-agi-01-frontier","choices":[{"index":0,"delta":{"content":"Hello"},"finish_reason":null}]}

data: {"id":"chatcmpl_8f0c2a…","object":"chat.completion.chunk","created":1781214488,"model":"lucidquery-agi-01-frontier","choices":[{"index":0,"delta":{},"finish_reason":"stop"}]}

data: [DONE]

By default no usage is sent (matching OpenAI). Opt in with stream_options: { "include_usage": true } to receive a final chunk whose choices is empty and whose usage is populated.

# with stream_options: { "include_usage": true }, a final chunk carries usage:
data: {"id":"chatcmpl_8f0c2a…","object":"chat.completion.chunk","model":"lucidquery-agi-01-frontier","choices":[],"usage":{"prompt_tokens":15,"completion_tokens":3,"total_tokens":18}}

data: [DONE]

A LucidQuery extension: set thinking: trueto receive the model's reasoning trace alongside its answer (most useful with lucidquery-agi-01-frontier). The trace is returned as reasoning_content on the message. include_reasoning and reasoning: { "enabled": true } are accepted aliases.

resp = client.chat.completions.create(
    model="lucidquery-agi-01-frontier",
    messages=[{"role": "user", "content": "Prove that sqrt(2) is irrational."}],
    extra_body={"thinking": True},
)

print(resp.choices[0].message.reasoning_content)  # the model's reasoning
print(resp.choices[0].message.content)            # the final answer

Every API response carries an id and your current rate-limit standing.

x-request-id: req_3f9a2c8b1d4e…
x-ratelimit-limit-requests: 100
x-ratelimit-remaining-requests: 99
x-ratelimit-limit-tokens: 1000000
x-ratelimit-remaining-tokens: 999985

Limits are enforced per API key over rolling windows. Exceeding one returns 429 with a rate_limit_error. The defaults below apply to new keys.

Errors use standard HTTP status codes and an OpenAI-shaped body: an error object with message, type, param, and code.

{
  "error": {
    "message": "API key is required.",
    "type": "authentication_error",
    "param": null,
    "code": "unauthorized"
  }
}

Returns the catalog of available model ids in the OpenAI list format.

curl https://api.lucidquery.com/v1/models \
  -H "Authorization: Bearer $LUCIDQUERY_API_KEY"