LLM API

The LLM API handles sessions and language model completions. It is the core interface for analytical Q&A within Lakehousecat, connecting users to the configured provider models (OpenAI, Anthropic, Google, Azure, AWS Bedrock).

Base URL: http://<host>:42015

What you can do

• Create and manage sessions
• Run streaming session completions against any configured provider model
• Search, archive, clone, and share sessions

Authentication

All endpoints require a Bearer token. Generate your API key in the Lakehousecat UI:

Account Settings → Security → API Keys → Generate API Key

Authorization: Bearer <your-api-key>

Quick Start

List available models

Provider and custom models are managed via the LHC Core API, not the LLM service:

curl -X GET "http://localhost:42021/api/v1/lhc/models" \
  -H "Authorization: Bearer <your-api-key>"

import requests

LHC_CORE = "http://localhost:42021"
HEADERS = {"Authorization": "Bearer <your-api-key>"}

response = requests.get(f"{LHC_CORE}/api/v1/lhc/models", headers=HEADERS)
models = response.json()

Create a new session

curl -X POST "http://localhost:42015/api/v1/llm/sessions/new" \
  -H "Authorization: Bearer <your-api-key>" \
  -H "Content-Type: application/json" \
  -d '{"title": "My Session"}'

BASE_URL = "http://localhost:42015"

response = requests.post(
    f"{BASE_URL}/api/v1/llm/sessions/new",
    json={"title": "My Session"},
    headers=HEADERS,
)
session = response.json()
session_id = session["id"]

Send a session completion

curl -X POST "http://localhost:42015/api/v1/llm/session/completions" \
  -H "Authorization: Bearer <your-api-key>" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4o",
    "messages": [
      {"role": "user", "content": "What is the total revenue this year?"}
    ]
  }'

response = requests.post(
    f"{BASE_URL}/api/v1/llm/session/completions",
    json={
        "model": "gpt-4o",
        "messages": [
            {"role": "user", "content": "What is the total revenue this year?"}
        ],
    },
    headers=HEADERS,
)
result = response.json()
print(result["choices"][0]["message"]["content"])

List your sessions

curl -X GET "http://localhost:42015/api/v1/llm/sessions/paginated?page=1&page_size=20" \
  -H "Authorization: Bearer <your-api-key>"

response = requests.get(
    f"{BASE_URL}/api/v1/llm/sessions/paginated",
    params={"page": 1, "page_size": 20},
    headers=HEADERS,
)
sessions = response.json()

Endpoint Groups

Resource	Path	Description
Session Completions	/api/v1/llm/session/completions	Run model completions within a session
Sessions	/api/v1/llm/sessions	Session lifecycle management (create, list, archive, clone, share)
Models	/api/v1/lhc/models	List all accessible provider and custom models (LHC Core API)

UI Equivalent

Sessions managed via this API correspond to the Sessions workspace in the Lakehousecat UI. Any session created or modified via API is immediately visible in the UI session list.