API key cursor quickstart guide

Authentication and base URL

Start by generating a key on the Get API key page. Click Continue with Google or use an email and password. The system shows your key immediately. Keep it secure.

Point your client to HTTPS://api.apikeycursor.com/v1. Pass the key in the Authorization header as a Bearer token. This base URL works with the official OpenAI SDKs and any compatible client. You only need one active key per account.

• Auth Header: Authorization: Bearer YOUR_KEY
• Base URL: HTTPS://api.apikeycursor.com/v1

Copycurl https://api.apikeycursor.com/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "uncensored",
    "messages": [{"role": "user", "content": "fetch current server load stats"}]
  }'

The chat completions endpoint

Call POST /v1/chat/completions to generate text. Send your messages in the standard JSON format. The model id is uncensored. It runs on our servers and does not refuse lawful adult topics.

Input tokens cost $0.15 per 1M. Output tokens cost $0.50 per 1M. Errors and refusals are free. Credit never expires. You can top up with USDT on TRC20 or USDC on Base. Start at $20. Get a 15% bonus at $200.

Copyfrom openai import OpenAI

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

resp = client.chat.completions.create(
    model="uncensored",
    messages=[{"role": "user", "content": "Summarise this thread without softening it."}],
)
print(resp.choices[0].message.content)

Streaming responses via SSE

Enable streaming by setting stream to true. You receive Server-Sent Events. Each chunk contains partial text. The final chunk includes token usage data.

Handle the stream in your code. Process each chunk as it arrives. This reduces perceived latency for long responses. The model supports a 100,000 token context window. Max output is 16,000 tokens per request.

Copyimport OpenAI from "openai";

const client = new OpenAI({ baseURL: "https://api.apikeycursor.com/v1", apiKey: process.env.API_KEY });

const resp = await client.chat.completions.create({
  model: "uncensored",
  messages: [{ role: "user", content: "Draft a villain monologue for my game." }],
});
console.log(resp.choices[0].message.content);

Function calling and JSON mode

Use the tools parameter for function calling. Define your functions in the schema. The model returns structured data. You can also force JSON output with response_format set to json_object.

This works for API key cursor integrations. Validate the JSON on your side. The model does not guarantee perfect syntax. Use this for automation. Parse the tool calls. Execute them. Send the results back.

Copystream = client.chat.completions.create(
    model="uncensored",
    messages=[{"role": "user", "content": "Tell the story in second person."}],
    stream=True,
)
for chunk in stream:
    if chunk.choices and chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="", flush=True)

Parameters: temperature and seed

Control creativity with temperature. Higher values make outputs more random. Lower values make them deterministic. Use seed for reproducible results. This helps debug prompt engineering.

Other parameters include top_p, stop, and presence_penalty. Adjust them based on your use case. Fine-tune the response style. The uncensored model responds well to direct prompts. It does not over-refuse controversial topics.

Check the pricing page for token costs. Use the trial credit to test.

Limits, errors, and context

You get 300 requests per minute. You can run 8 requests simultaneously. The request body limit is 8 MB. The context window is 100,000 tokens total. Prompt and completion share this space.

Errors return standard HTTP codes. 401 means invalid key. 402 means no credit. 429 means rate limit exceeded. Check the Support page for refunds. Mistakes like double charges are fixed there. Credit is not refunded otherwise.