Quickstart for Grok Unrestricted users
Connect your existing OpenAI-compatible client to our uncensored model in minutes. Replace the base URL and API key, then send requests to the same endpoints you already know.
Prerequisites
Before sending requests, ensure you have an API key and a client configured for OpenAI-compatible endpoints. Create an account on our site using Continue with Google or an email and password. Your API key is displayed immediately and is the only credential you need. No phone number or credit card is required for the initial setup. You will receive a $0.50 trial credit valid for 7 days to test the endpoint. Claim this credit with one click on your API key page. For production use, top up with USDT (TRC20) or USDC (Base). The service supports standard text-in, text-out workflows with a 100,000-token context window.
Authentication
Include your API key in the Authorization header of every request. Use the Bearer scheme followed by your key string. This authenticates your account against the api.grokunrestricted.com base URL. If the key is invalid or missing, the API returns an error indicating authentication failure. Ensure your key is current; generating a new key replaces the previous one. No additional scopes or permissions are required for standard chat completions.
Chat Completions Endpoint
Send messages to the POST /v1/chat/completions endpoint. Specify the model ID as uncensored. The request body should include the messages array with your conversation history. The API returns a text response within the choices array. This endpoint supports standard parameters like temperature, top_p, and stop. Requests are charged based on actual token usage. Errors and refusals do not incur charges.
Streaming Responses
Use Server-Sent Events (SSE) for streaming. Setstream to true in your request. The API returns a stream of chunks, each containing partial text. The final chunk includes the total token usage for the request. This approach reduces perceived latency for large responses. Use the official SDKs to handle stream parsing automatically. Streaming works with all standard parameters.
Function Calling
Enable tool use by providing a tools array in the request. Define function signatures with names, descriptions, and parameter schemas. The model returns tool calls in the response when appropriate. You must execute the tool on your side and pass the results back in subsequent messages. The API does not validate tool output accuracy; you are responsible for handling the results. Use tool_choice to control how the model selects functions.
JSON Mode
Force the model to output valid JSON by setting response_format to {"type": "json_object"}. This is useful for structured data extraction or configuration generation. The model will attempt to conform to JSON syntax rules. Ensure your prompt clearly describes the expected structure. Invalid JSON responses may still occur if the model struggles with complex schemas. Validate the output on your client side before parsing.
cURL
curl https://api.grokunrestricted.com/v1/chat/completions \
-H "Authorization: Bearer $API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "uncensored",
"messages": [{"role": "user", "content": "Write a blunt product review of a cheap VPN."}]
}'Python
from openai import OpenAI
client = OpenAI(base_url="https://api.grokunrestricted.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)Node.js
import OpenAI from "openai";
const client = new OpenAI({ baseURL: "https://api.grokunrestricted.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);Streaming
stream = 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)Under the hood: specs
Everything the endpoint can and cannot do, in one place — check it before you top up.
| Spec | Value |
|---|---|
| API format | OpenAI Chat Completions schema; official openai SDKs work unchanged |
| Model ID | uncensored |
| Endpoints | POST /v1/chat/completions · GET /v1/models |
| Base URL | https://api.grokunrestricted.com/v1 |
| Authentication | Bearer token in the Authorization header |
| Sampling parameters | temperature, top_p, stop, seed and the two penalties are passed through |
| Context window | 100,000 tokens, input and output combined |
| Streaming | Yes — server-sent events; the last chunk carries token usage |
| Function calling | Yes — tools, tool_choice; replies carry tool_calls, also when streaming; send results back as role: tool |
| Max output | prompt + completion fit within 100,000 tokens; max_tokens optional, no separate output cap |
| Structured output | response_format: {"type": "json_object"} |
| Rate limit | 300/min per key |
| Parallel requests | up to 8 in parallel per key |
| Headers | X-Request-Id, X-Balance-USD, X-RateLimit-Limit-Requests, X-RateLimit-Limit-Concurrency |
| Max body | up to 8 MB per request |
| Free trial | $0.50 of credit valid 7 days, no card needed · Trial key: 2 parallel requests, 60 req/min; full limits (8 and 300) after first top-up |
| Credit expiry | paid credit never expires, no subscription |
| Bonus credit | +5% from $50, +10% from $100, +15% from $200 |
| Payment | USDT (TRC20) or USDC (Base), any whole amount from $20 to $500 |
| How you pay | prepaid credit, charged by real token usage; errors and refusals are free |
| Price | input $0.15 / 1M tokens, output $0.50 / 1M tokens |
| Account | sign in with Google or with e-mail + password |
| Keys | one key per account, regenerate any time (the old one stops working) |
| Content policy | adult content allowed; sexual content involving minors is refused |
Error codes
The type field is stable, the message is for humans. Errors cost nothing.
| Code | Type | Meaning |
|---|---|---|
400 | bad_request | invalid JSON, empty messages, bad parameter, or a prompt longer than the window — fix and resend |
401 | missing_key · invalid_key · key_revoked | check the Authorization header or use your current key |
402 | no_credit | out of credit; add credit and retry |
403 | content_blocked | sexual content involving minors — refused, not billed |
404 | not_found | unknown endpoint |
413 | request_too_large | body over 8 MB |
429 | rate_limited · concurrency | over 300/min or 8 parallel — back off and retry |
503 | upstream_busy | model busy — retry in a few seconds |
Questions and answers
How do I handle rate limits?
The API allows 300 requests per minute per key and 8 concurrent requests. If you exceed these limits, the API returns an error. You should implement retry logic with exponential backoff in your client to handle temporary overload efficiently.
What happens if I run out of credit?
Requests will fail if your account balance is insufficient. Errors are free, so you will not be charged for failed requests due to lack of funds. Top up your account with crypto to resume service. Credit never expires.
Is the context window 100k tokens total?
Yes, the context window is 100,000 tokens for both prompt and completion combined. There is no separate output cap. If you do not specify max_tokens, the response can run to the end of the window.
Your key is one form away
Create an account, copy the key, change the base URL. That is the whole setup.
Get API key