EN ▾
Get API key

Uncensored coding AI API

Uncensored Cursor API alternative

A single, open-weight model optimized for coding workflows. Drop it into Cursor as a drop-in alternative to standard providers.

  • 100k context window
  • No content refusals
  • $0.15 per 1M tokens

Get API keyRead the docs

Try it in one request

https://api.cursorapikey.com/v1
curl https://api.cursorapikey.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."}]
  }'
01

What you get

OpenAI-Compatible

Works with official SDKs. Change your base URL and API key to use our endpoint immediately.

Uncensored Model

Our open-weight model does not refuse lawful adult, fictional, or security-research topics.

100k Context Window

Handle large codebases with a 100,000 token limit for both prompt and completion.

Streaming Support

Get real-time responses via Server-Sent Events for a better developer experience.

Function Calling

Use tool calling to integrate the model into your automated coding agents.

Simple Pricing

Pay-as-you-go prepaid credit. No monthly fees or subscriptions required.

02

How it works

  1. Sign Up

    Create an account with just an email and password.

  2. Get Your Key

    Receive your API key immediately after signing up.

  3. Integrate

    Update your Cursor settings to use our base URL and key.

03

What people build with it

Code Generation

Generate code without context refusals for standard programming tasks. Works well for creating functions and classes.

Debugging

Paste error logs and code snippets for analysis. The model focuses on solving the technical problem.

Refactoring

Request code improvements or style changes. Get consistent results without generic polite refusals.

Documentation

Generate comments and docstrings for your codebase. The model understands code context deeply.

What is the Cursor API?

Cursor is a popular AI code editor that uses large language models to assist developers. The standard version often relies on proprietary models that may refuse certain prompts based on content policies. Our service provides a dedicated API endpoint that serves a single, uncensored large language model. This model is open-weight and hosted on our own GPU servers. It is designed specifically for coding contexts where developers need consistent, refusal-free assistance.

We are not Anthropic, OpenAI, or Cursor. We are an independent provider offering a raw, text-based API. You send text in, and get text out. There are no embeddings, image, audio, or video generation capabilities. This simplicity allows for lower costs and faster iteration for coding tasks. You can use this API with any OpenAI-compatible client or SDK.

Why Choose an Uncensored Model?

Standard coding models often refuse to generate code for security vulnerabilities or adult themes in fiction. This can interrupt your workflow when you need a direct answer. Our uncensored model does not apply these content refusals for lawful adult use. It will generate the code or explanation you ask for, provided it is not sexually explicit content involving minors.

This is particularly useful for security researchers and developers working on sensitive codebases. You get the raw output without the model deciding it is 'unsafe' to answer. You can test edge cases, generate verbose explanations, and work with controversial topics without hitting a content block. The model is tuned to prioritize answering your coding questions directly.

OpenAI-Compatible Integration

Our API follows the OpenAI Chat Completions standard. This means you can use it with Cursor, VS Code extensions, or any tool that supports OpenAI-compatible endpoints. You only need to change two things: your base URL and your API key.

Base URL: https://api.cursorapikey.com/v1
Model ID: uncensored

Streaming and function calling are fully supported. This allows you to build custom agents or get real-time feedback in your editor. The model ID is always 'uncensored'. You do not need to switch between different model versions. Everything is handled through a single, consistent endpoint.

Simple Pay-As-You-Go Pricing

We charge based on token usage. There are no monthly subscriptions or hidden fees. You prepay for credits, and they never expire. The cost is $0.15 per 1 million input tokens and $0.50 per 1 million output tokens. This is significantly cheaper than many major providers for high-volume coding tasks.

You can top up by crypto (USDT on TRC20, USDC on Base). Starting at $20, you get bonus credits for larger purchases. A $50 top-up gives you +5% bonus credit. A $100 top-up gives you +10% bonus credit. A $200 top-up gives you +15% bonus credit. New accounts get $0.50 in trial credit for 7 days. Claim this credit with one click on your API key page. No credit card is needed for the trial.

04

Questions and answers

Is this the official Cursor API?

No. We are an independent service. Cursor uses various models for its editor. Our API is a drop-in alternative that you can configure within Cursor or any other OpenAI-compatible client. We do not claim to be affiliated with Cursor Software Inc.

What is the context window size?

Our model supports a 100,000 token context window. This includes both the prompt and the completion. This is sufficient for most large codebases and complex coding tasks. It is larger than the standard 8k context used by some legacy models.

Does the API support streaming?

Yes. You can enable streaming by setting the stream parameter to true in your requests. This returns Server-Sent Events (SSE) for real-time token delivery. This is ideal for coding agents that need to display output as it is generated.

Is my data used for training?

No. We do not use your prompts or completions for training our model. Your data is private to your account. We only store what is necessary to process your requests and manage your billing. You can regenerate your API key at any time to revoke access for old keys.

Who is this NOT for?

This API is not for users who need image generation, audio processing, or embeddings. We do not offer fine-tuning or model routing. If you need multiple models or specialized capabilities like vision, this is not the right choice. We offer a single, uncensored text model for coding and general text tasks.

Your key is one form away

Create an account, copy the key, change the base URL. That is the whole setup.