← Developers

Developer Docs

One OpenAI-compatible API for everything List Cloud. Base URL: https://api.listcloud.com/v1

Quickstart

Create a key in your workspace, then call the API. Replace YOUR_KEY with a real key.

curl https://api.listcloud.com/v1/chat/completions -H "Authorization: Bearer YOUR_KEY" -H "Content-Type: application/json" -d '{"model":"List-Loop-1.0-Brain","messages":[{"role":"user","content":"hi"}],"max_tokens":50}'

Copy curl

from openai import OpenAI
client = OpenAI(base_url="https://api.listcloud.com/v1", api_key="YOUR_KEY")
r = client.chat.completions.create(model="List-Loop-1.0-Brain", messages=[{"role": "user", "content": "hi"}])
print(r.choices[0].message.content)

Copy Python

import OpenAI from "openai";
const client = new OpenAI({ baseURL: "https://api.listcloud.com/v1", apiKey: "YOUR_KEY" });
console.log((await client.chat.completions.create({ model: "List-Loop-1.0-Brain", messages: [{ role: "user", content: "hi" }] })).choices[0].message.content);

Copy JavaScript

Authentication

Send your key as a Bearer token: Authorization: Bearer YOUR_KEY. Workspace keys start with lc-clor- and carry a credit budget; usage is metered per call and blocked when the budget runs out.

Models

GET /v1/models lists what your key can use. The flagship chat model is List-Loop-1.0-Brain: chat, reasoning, code, and tool calling.

curl https://api.listcloud.com/v1/models -H "Authorization: Bearer YOUR_KEY"

Copy

Chat completions

POST /v1/chat/completions with model, messages, and optional max_tokens, temperature, stream. Add "stream": true for server-sent events. Legacy POST /v1/completions (prompt-based) is also accepted. Embeddings are not supported by chat models and return a 400 error.

Reasoning effort

Default calls are fast with no extra reasoning. For harder problems, add a reasoning hint in the JSON body or the prompt:

{"model":"List-Loop-1.0-Brain","messages":[{"role":"user","content":"solve a hard equation"}],"reasoning":{"effort":"high"}}

Copy

Tools, Cursor & MCP

Function calling uses the OpenAI tools format with tool_choice: "auto". In Cursor set the OpenAI base to https://api.listcloud.com/v1 with model List-Loop-1.0-Brain. MCP-compatible clients can point at the same base URL with your key.

Errors & limits

Errors come back as JSON with a List Cloud provider tag and HTTP status (400 bad request, 401 bad key, 402 budget exhausted, 429 rate limited, 5xx retry with backoff). Usage is metered against your key budget — check spend in the workspace console.

Health

GET /health returns service status without authentication. Try the live calls on the Demo page.