Standard Integration
Use the official OpenAI SDKs or any compatible client. Change the base URL and API key to start sending requests to our uncensored model.
unfilteredaiapi.comhttps://api.unfilteredaiapi.com/v1
Unfiltered AIRaw, uncensored LLM API for developers.
Uncensored LLM API
Use supplied text as the basis for creative revisions and contrasting viewpoints. Keep source facts separate from fictional claims, then review what the model changes or adds.
OpenAI-compatible endpoint
$0.15/1M input tokens
No subscription required
Try it in one request
curl https://api.unfilteredaiapi.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."}]
}'from openai import OpenAI
client = OpenAI(base_url="https://api.unfilteredaiapi.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)import OpenAI from "openai";
const client = new OpenAI({ baseURL: "https://api.unfilteredaiapi.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);Use the official OpenAI SDKs or any compatible client. Change the base URL and API key to start sending requests to our uncensored model.
Use the model within the service’s content rules and review each generated reply. Sexual content involving minors is prohibited, and other requests can still be refused.
Input and output together must fit within 100,000 tokens. Reserve response space and send only the history needed for the current task.
Pay-as-you-go with no monthly fees. Credits never expire. Top up from $20 or $50 to receive bonus credit.
Support for Server-Sent Events (SSE) streaming and function calling. Build dynamic, real-time applications with full control over the output.
We run our own open-weight model on dedicated GPU servers. No aggregators, no hidden routing, and no model switching behind the scenes.
Sign up with an email and password to receive your API key instantly.
Add funds by crypto (USDT on TRC20, USDC on Base). Start with the $0.50 free trial if you prefer. Claim this credit with one click on your API key page.
Send POST requests to /v1/chat/completions using your preferred OpenAI-compatible SDK.
Send character details and recent dialogue, then compare the new reply with the stored scene. Keep refusal handling and content review in the application.
Produce unrestricted stories and scripts. The model does not sanitize edgy or unconventional narrative elements.
Create a small synthetic test batch with explicit labels and output fields. Check every record before using it in a dataset; generated text is not verified evidence.
Analyze controversial topics or security research data without the model refusing to discuss sensitive subjects.
Use the endpoint for bounded text-generation tasks such as revising supplied dialogue. Keep source facts, requested changes and content limits explicit, then inspect the output. The service supplies one model ID, not a guarantee of fixed behavior.
We do not route your traffic through multiple vendors or obscure the model provenance. You are interacting directly with our uncensored llm api instance. The only hard limit is that sexual content involving minors is always blocked, as is standard for most models.
Drop-in compatibility means you can integrate our API into existing workflows with minimal code changes. Use the standard /v1/chat/completions endpoint with the official OpenAI SDKs or any client that supports OpenAI-compatible endpoints. Streaming via SSE and tool/function calling are fully supported.
The base URL is https://api.unfilteredaiapi.com/v1. Simply update your base_url and API key in your configuration. This approach avoids the complexity of managing multiple model endpoints or dealing with aggregator-specific quirks.
Our model supports a 100,000 token context window, covering both the prompt and the completion. This allows for long-running roleplay sessions, detailed creative writing, or processing substantial documents in a single request. Maintaining context over extended interactions is critical for immersive applications.
Unlike smaller context windows that force truncation or summarization, this capacity ensures you retain the full narrative thread. It is particularly useful for applications that require deep state tracking or long-form content generation without frequent resets.
We charge $0.15 per 1M input tokens and $0.50 per 1M output tokens. There are no monthly subscriptions or hidden fees. You pay only for what you consume, and your paid credit does not expire. This model allows you to scale usage purely based on your actual token consumption. Trial credit has a separate 7-day validity period.
Top-ups start at $20, with bonus credits available for larger deposits. You can pay by crypto (USDT or USDC). This transparency ensures you know exactly what you are paying for, without being locked into a tier that may not match your usage patterns.
We run a single open-weight model on our own GPU servers. It is not GPT, Claude, Gemini or Grok. It is a distinct model tuned for uncensored output.
You can test fictional adult-character prompts within the service’s rules. The model may refuse or produce unsuitable output. Sexual content involving minors is prohibited.
No. We only offer text chat completions. There are no embeddings, image, audio, or video generation capabilities. We focus on raw text output.
Create an account and obtain the current key. Eligible accounts can request a $0.50 trial for 7 days; verification and processing may be required. Check the balance before testing.
You can make 300 requests per minute per API key. The maximum request body size is 8 MB. Each account has one key, which can be regenerated at any time.
Create an account, copy the key, change the base URL. That is the whole setup.
Supply two fictional speakers’ statements about a disputed city-planning proposal and a separate list of facts allowed in the scene. Request one shortened statement per speaker and a JSON list of claims not supported by those facts. Compare each rewrite with the original position and check the flagged claims manually. This is an editorial test with synthetic material. The model does not verify real-world evidence or guarantee that a rewritten argument is accurate.
Practical articles on setup, pricing and integration.
Free Unfiltered AI API: Real Options and Their LimitsLooking for a free unfiltered AI API? See what free access really gives you, where it breaks, and what $0.50 of trial credit and pay-per-token buy…
Node JS AI API: Streaming, Tools and Errors in ExpressBuild a Node JS AI API with the OpenAI SDK: change the base URL, stream replies over SSE, call tools and handle 429 errors in an Express route,…
LLM API Response Empty? Causes and FixesGetting an empty reply from a chat completions call? Check finish_reason, tool calls, streaming chunks, max_tokens, stop sequences and 429s. Code…
Unfiltered AI for game NPC dialogue and interactive fictionUse the unfiltered LLM API for game NPCs and interactive fiction. OpenAI compatible, pay per token. Start building now.
An unfiltered chatbot backend with FastAPI and server-sent eventsBuild an unfiltered chatbot backend using FastAPI and SSE. Stream tokens via StreamingResponse for OpenAI compatible API integration. Start coding…
Unfiltered AI API: frequently asked questionsUnfiltered LLM API FAQ: access, billing, and limits. OpenAI compatible API details for streaming chat platforms. Read more.