Uncensored AI API free trial and FAQ
Eligible new accounts can request $0.50 of trial credit, valid for 7 days, without supplying payment details. Use the claim control on the API key page, complete email confirmation if requested, and check the balance after processing. This guide then shows how to send a first request, estimate token costs and handle an expired or exhausted trial.

Key points
- The free trial is $0.50 credit for new accounts, valid 7 days; sign up with email and password and the key appears immediately. Claim this credit with one click on your API key page.
- $0.50 covers 2M input tokens or 500K output tokens; a typical 1,000-in, 500-out request costs $0.00075, about 666 requests.
- After the trial or when credit is used up, requests return 402 no_credit until you top up; prepaid balance never expires.
- Lawful adult content is not refused; sexual content involving minors is always blocked with a 403. Prompts are not used for training.
How the uncensored AI API free trial works
New accounts get $0.50 of free credit, valid for 7 days, with no payment details required. It is a normal account with a normal key, so anything that works on a paid balance works during the trial. Claim this credit with one click on your API key page. Here is the full path from nothing to a working call.
- Sign up. Open the get-api-key page and register with an email address and a password.
- Copy your key. The key is shown immediately after sign-up. You get one key per account. Store it in an environment variable called
API_KEY, not in source code. - Make the first call. Send a request to
POST https://api.uncensoredaiapis.com/v1/chat/completionswith the headerAuthorization: Bearer <key>and modeluncensored. - Check usage. Each response includes a
usageobject with prompt and completion tokens, so you can see exactly what each call consumed. - Decide before day 7. The trial credit expires after 7 days. After that, or once the credit is used up, requests return 402
no_credituntil you top up your prepaid balance.
export API_KEY="paste-your-key-here"
curl https://api.uncensoredaiapis.com/v1/chat/completions \
-H "Authorization: Bearer $API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "uncensored",
"messages": [{"role": "user", "content": "Say hello in one sentence."}],
"max_tokens": 50
}'If you prefer an SDK, the OpenAI Python and Node packages work by setting base_url to the same address; ready-to-run snippets are on the code examples page. The trial has no feature restrictions described beyond the credit amount and the 7-day window, so it is a fair way to check output quality, streaming and tool calling against your real prompts before you spend anything.
What $0.50 buys in tokens
Pricing is $0.15 per million input tokens and $0.50 per million output tokens. Your $0.50 trial credit therefore covers 2,000,000 input tokens if you only send prompts, or 500,000 output tokens if you only count replies. Claim this credit with one click on your API key page. Real requests mix both, so the useful number is requests per $0.50. The table works through common shapes; each row is input cost plus output cost.
| Request shape | Cost per request | Requests for $0.50 |
|---|---|---|
| Quick answer: 200 in, 200 out | $0.00005 + $0.0002 = $0.00025 | 2,000 |
| Chat turn: 2,000 in, 300 out | $0.0005 + $0.0003 = $0.0008 | 625 |
| Standard: 1,000 in, 500 out | $0.00025 + $0.0005 = $0.00075 | about 666 |
| Long scene: 6,000 in, 2,700 out | $0.0015 + $0.0027 = $0.0042 | about 119 |
| Maximum reply: up to the end of the 100,000-token window | $0.00025 + $0.016 = about $0.0163 | about 30 |
In plain terms, 500,000 output tokens is roughly 370,000 words of generated text if you assume about three-quarters of a word per token, which is a rough rule of thumb for English rather than an exact figure. That is several novels of output, or many hundreds of chat turns. Two habits stretch the trial further: set max_tokens on every call, and keep conversation history short, since history is input and is re-sent each turn. For a deeper look at per-task costs, see the use cases page, or the pricing page for rates.
Account and key questions
Do I need to enter payment details to start the free trial?
No. You sign up with an email address and a password, and the $0.50 credit is applied to the new account without payment details.
When do I get my API key?
Immediately. The key is displayed as soon as the account is created on the get-api-key page.
How many keys can I have?
One key per account. If you need to rotate it, regenerate it from your account; the old key is revoked instantly, so update every deployment that uses it straight away.
What if my key leaks?
Regenerate it. Because the old key stops working the moment you do, a leaked key cannot be used afterwards. Keep the key in a server-side environment variable and never include it in browser code or public repositories.
Can I run several apps on one account?
They would share the one key, and the rate limit of 300 requests per minute applies to that key. If apps are independent, plan their traffic with that shared limit in mind.
Who gets the free trial?
The $0.50 credit is for new accounts and is valid for 7 days. Use it to test the service before you decide whether to top up a prepaid balance.
Billing questions
How does billing work?
It is prepaid. You top up a balance and each request deducts input and output tokens at $0.15 and $0.50 per million. There is no subscription and no monthly minimum.
Does my balance expire?
Prepaid balance never expires. The trial credit is different: it is valid for 7 days from sign-up.
What happens when the credit runs out?
Requests return HTTP 402 with the code no_credit. That also happens if the trial expires before you use the credit. Top up the prepaid balance and the same key works again.
How do I estimate cost before sending a request?
Multiply expected input tokens by $0.15 per million and expected output tokens by $0.50 per million. Setting max_tokens gives you a hard ceiling on the output part. The pricing page has worked examples.
Do streaming requests cost more?
No. Streaming returns the same tokens as a normal request, so the price is identical. A final usage chunk tells you what the call consumed.
API and parameter questions
Is the API compatible with the OpenAI SDK?
Yes. Use the official openai package in Python (v1+) or Node (v4+), set the base URL to https://api.uncensoredaiapis.com/v1, and pass model uncensored.
Which endpoints exist?
Two: POST /v1/chat/completions and GET /v1/models. Anything else returns 404.
What is the context window?
100,000 tokens, shared between the prompt and the completion. If the prompt tokens alone exceed that, the API returns a 400 error; if prompt tokens plus max_tokens exceed it, the reply simply stops at the end of the window.
What are the max_tokens defaults and limits?
If max_tokens is not set, the response can run to the end of the window. Request bodies can be up to 8 MB.
Does it support streaming and function calling?
Yes to both. Set stream: true for server-sent events, with a final usage chunk added automatically, and use tools and tool_choice in the OpenAI format for function calling.
Can I generate images, audio or embeddings?
No. The API is text only: no embeddings, images, audio, video or fine-tuning, and there is a single model.
Error questions
Errors return JSON in the form {"error":{"code":...,"message":...}}. The HTTP status tells you the category; the code and message give detail.
What does a 401 mean?
The key is missing or invalid. Check the Authorization: Bearer <key> header and make sure you did not paste an old key after regenerating.
What does 402 no_credit mean?
Your balance is used up or the free trial expired. Top up the prepaid balance, then retry.
What does 403 content_blocked mean?
The request was blocked by the content rule that always applies: sexual content involving minors, including fiction and roleplay. Retrying the same request will not change the result.
Why do I get 400?
The request is malformed (for example an invalid max_tokens value), or the prompt alone exceeds the 100,000-token context. Lower max_tokens or trim the history.
What do 429 and 503 mean, and should I retry?
429 means you passed 300 requests per minute on that key, and 503 upstream_busy means to retry in a few seconds. Retry both with exponential backoff and jitter; the code examples include a working loop.
Why do I get 404?
You called an endpoint that does not exist. Only /v1/chat/completions and /v1/models are available, and the base URL must end in /v1.
Content rules questions
What content is allowed?
Lawful adult content, fiction and controversial topics are not refused. The service is intended for adults and lawful use.
What is always blocked?
Sexual content involving minors. It returns 403 content_blocked, including when it appears in fiction or roleplay, and no prompt or setting changes this.
Do I need to verify my users are adults?
If your product offers adult content, yes. You are responsible for age gating in your own app and for following the laws that apply to your users.
Is this a good fit for any audience?
The API is for adults. It is not intended for products aimed at children or at schools.
Privacy questions
Are my prompts used for training?
No. Prompts are not used for training.
How should I handle my API key and user data in my own app?
Call the API from your backend so the key is never exposed to browsers. Send only the data a request needs, and apply the privacy policy that suits your own users, since your app decides what ends up in prompts.
Can I tell my users what I use?
Yes. You can describe in your own privacy notice that your app sends prompts to a third-party text-generation API, and state the training commitment above if it is relevant.
Ready to test? Create an account and use the first-call snippet above. Return to the homepage for an overview of the API.
Questions and answers
How does the uncensored AI API free trial work?
Sign up with an email and password, copy the key shown immediately, and call POST /v1/chat/completions with model uncensored. New accounts get $0.50 credit, valid 7 days, with no payment details needed.
What does the $0.50 free trial credit buy?
At $0.15 per million input and $0.50 per million output tokens, $0.50 covers 2 million input tokens or 500,000 output tokens. A request with 1,000 input and 500 output tokens costs $0.00075, so about 666 requests.
What happens when the free trial ends?
Requests return 402 with the code no_credit once the credit is used or the 7 days pass. Top up the prepaid balance and the same key works again. Prepaid balance never expires.
Is anything blocked on this uncensored AI API?
Lawful adult content, fiction and controversial topics are not refused. Sexual content involving minors is always blocked with a 403 content_blocked error, including fiction and roleplay.
Are my prompts used for training?
No. Prompts are not used for training.
Your key is one form away
Create an account, copy the key, change the base URL. That is the whole setup.