EN ▾
Get API key

Uncensored LLM API

Uncensored LLM API for Unrestricted AI

Send bounded text-processing tasks through a Chat Completions endpoint. Define the allowed labels and output fields, validate every returned record, and estimate cost from actual token usage.

Get API keyRead the docs

Try it in one request

curl https://api.uncensoredendpoint.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."}]
  }'

What you get

Standard Interface

Use the familiar POST /v1/chat/completions endpoint with any OpenAI-compatible client. Streaming, tool calling, and functions work exactly as documented.

Uncensored Model

The model generates text subject to the service’s content rules and its own behavior. Review returned answers and handle refusals or failures in the application.

Private by Design

Your prompts are not used for training. We keep your data separate from the training pipeline, ensuring your context remains yours.

Transparent Pricing

Pay-as-you-go credits that never expire. No monthly fees or subscription traps. Just clear per-token costs for input and output.

High Context Window

Reserve response space within the combined 100,000-token context limit. Send only the document passages needed for the current question and verify the answer against them.

Simple Limits

300 requests per minute per key with an 8 MB request body limit. Easy to monitor and scale for most applications.

How it works

Sign Up

Create an account, obtain the current key and check available credit before making requests.

Add Credit

After a paid top-up, wait for processing and check the available balance before sending requests. Checkout shows the supported payment methods.

Start Coding

Point your OpenAI SDK to our base URL and start sending requests.

What people build with it

Uncensored Chatbots

Provide a character brief and recent dialogue. Compare the generated line with names, events and allowed actions before advancing the scene.

Content Generation

Provide an outline and explicit constraints for a writing draft. Check generated text for unsupported additions and handle refused or incomplete responses before using it.

Research & Analysis

Ask a bounded question about supplied document passages, then check the answer against those passages. Context capacity does not guarantee accurate or complete analysis.

Developer Testing

Test your application logic with a reliable, standard-compliant endpoint. Use the free trial credit to verify your integration before paying.

What is an Uncensored OpenAI-Compatible Endpoint?

An uncensored openai-compatible endpoint accepts the Chat Completions request format while serving an independent model. Requests use POST /v1/chat/completions. Validate the methods and parameters your client needs; compatibility does not imply identical answers or support for every OpenAI endpoint.

Define the task and validate the output

Specify a bounded task and review the returned text for accuracy and compliance with your constraints. Requests can be refused, fail or produce unsuitable output. Sexual content involving minors is prohibited.

Private by Design

Since this is an uncensored llm api built on the OpenAI standard, integration is straightforward. You simply change the base URL to https://api.uncensoredendpoint.com/v1 and provide your API key. The endpoint supports streaming via Server-Sent Events (SSE) and tool/function calling, just like the industry standard. You can use the official OpenAI SDKs for Python, Node.js, or any other language. The model ID is simply uncensored. This means you do not need to learn a new protocol or parse custom JSON structures. If it works with OpenAI, it works here. This standardization reduces development time and allows you to switch between endpoints if your needs change in the future.

Private by Design

The service states that prompts are not used for training. This does not specify every retention, access or logging practice. Read the applicable terms and omit information the task does not require before sending sensitive material.

Developer-First Pricing

Usage costs $0.15 per 1M input tokens and $0.50 per 1M output tokens. Paid credit does not expire; trial credit has a separate validity period. Paid deposits range from $20 to $500, with bonuses of 5% from $50, 10% from $100 and 15% from $200. Estimate spend from actual token counts.

Start Using the API

Check whether a text-only endpoint meets your application’s requirements; image generation, embeddings, audio processing and fine-tuning are not provided. Eligible accounts can request $0.50 in trial credit for 7 days, subject to verification and processing. Use model uncensored after checking the available balance.

Questions and answers

Is this the same as OpenAI's API?

It uses the same standard interface (OpenAI-compatible), but the underlying model is different. We host our own uncensored model on our own servers. It is not GPT-4 or GPT-3.5, nor is it a resell of another vendor's model. You get a different behavior profile, specifically tuned for fewer refusals.

Do you use my prompts for training?

No. Your prompts are not used to train our models. We keep your data separate from our training pipeline. This ensures that your creative work or sensitive context remains private and is not repurposed by us.

What is the context window size?

The model supports a 100,000 token context window. This includes both the input prompt and the generated output. This allows for long documents, extensive codebases, or multi-turn conversations with significant history.

Can I use this for NSFW content?

Use the model for tasks permitted by the service’s content rules. It may refuse or return unsuitable text. Sexual content involving minors is prohibited; keep output review and failure handling in the application.

How do I get started?

Obtain the current key from your account. Eligible accounts can request $0.50 of trial credit for 7 days; verification and processing may be required. Check the available balance before testing, and use the checkout’s listed methods for paid top-ups.

Your key is one form away

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

Get API keyRead the docs

Example: extract labels from a small synthetic batch

Send five invented support messages and ask for a JSON array with the original message_id and one label from account, billing or integration. Check that all five IDs appear once, every label is allowed and no personal details were added. Route uncertain items to manual review. Record actual input and output token usage when estimating cost. This example demonstrates a bounded text-processing task; it does not claim measured classification accuracy.