OpenAI Compatible
Drop in /v1/chat/completions with zero code changes.
Hosted Inference Layer
Uncensored local ai API for Developers
Connect your OpenAI SDK to a single uncensored local AI model with 100,000 token context and transparent crypto pricing.
- POST /v1/chat/completions endpoint.
- 100k token context window.
- USDT or USDC payments.
Test with one call
curl https://api.uncensoredlocalai.com/v1/chat/completions \
-H "Authorization: Bearer $API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "uncensored",
"messages": [{"role": "user", "content": "Generate a 500 word essay on the history of cryptography"}]
}'from openai import OpenAI
client = OpenAI(base_url="https://api.uncensoredlocalai.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.uncensoredlocalai.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);$0.15per 1M input tokens
$0.50Output tokens / 1M
100,000token context
$0.50trial credit
300requests per minute
Core capabilities
Uncensored Output
No refusals for lawful adult or controversial topics.
Streaming Support
SSE streaming with token usage in final chunk.
Function Calling
Native tools and tool_choice parameter support.
JSON Mode
Force structured output via response_format json_object.
Crypto Pricing
Pay per token with no monthly subscription fees.
Setup workflow
Generate Key
Sign up with Google or email to get your key.
Update Base URL
Set the base_url to our hosted inference endpoint.
Send Request
Call the model with id uncensored via your SDK.
What you can build
NSFW Roleplay
Run unrestricted character chats without content filter interruptions.
Raw Inference
Get direct model outputs without safety layer noise.
Content Generation
Produce unrestricted creative writing for adult audiences.
Dev Tools
Integrate uncensored logic into your own applications.
Switch your base_url in three lines of code
Update your base_url to point to our uncensored local AI endpoint and keep using the same OpenAI SDK clients. Replace the API key with your new credential, then set the model ID to "uncensored". This single configuration change routes your requests directly to our inference layer. Your existing code logic remains intact while gaining access to an unrestricted model that skips standard refusals. It takes seconds to verify connectivity with a simple chat completion call.
Built for NSFW roleplay, writing, and raw inference; not for enterprise SLAs
This uncensored LLM API powers direct content generation without refusal filters. Developers use it for unrestricted roleplay, raw creative writing, and niche dev tools where standard models block specific prompts. We do not offer SLA guarantees or multi-model routing. Use our endpoint for high-volume, uncensored inference. Skip this service if you require strict enterprise compliance or guaranteed uptime.
Uncensored local AI use cases: roleplay, content generation, and dev tools
Our uncensored local AI API serves roleplay, content generation, and dev tools by removing refusal layers from standard inference.
NSFW Roleplay
Send characters with custom personas. The model answers without filtering adult or controversial topics, ensuring consistent character voice for unlimited conversations.
Content Generation
Generate raw text for blogs, scripts, or marketing. Bypass safety filters that often block creative or edgy narratives in standard models.
Dev Tools
Integrate directly into your application. Use the standard OpenAI-compatible endpoint to build custom interfaces without managing local hardware.
A 6-point checklist to verify true uncensorship and API stability
Verify your uncensored LLM API provider with these six critical checks before committing budget.
- Hard Limit: Confirm explicit refusals only for minor content, not creative themes.
- Context Window: Ensure support for 100,000 tokens with clear 16,000 output limits.
- Pricing Transparency: Verify usage-based crypto billing with no hidden monthly fees.
- Rate Limits: Check for 300 requests per minute per key.
- Privacy: Confirm prompts are not used for training.
- SDK Compatibility: Validate standard OpenAI endpoints like /v1/chat/completions.
Avoid common pitfalls: context limits, and rate limits
Context limits, model hallucinations, and rate limits define your uncensored LLM API reliability. You must manage the 100,000 token window carefully. Requests exceeding the 8 MB body or 8 concurrent key limit trigger immediate errors. Stream responses via SSE to handle latency gracefully. Set explicit max_tokens to prevent runaway output costs. The model does not hallucinate less than others; it simply refuses fewer topics. Verify your integration handles these constraints before scaling.
From signup to first token: your 4-step integration guide
- Visit the Get API key page and authenticate via Google or email.
- Receive your single active key immediately; no phone number required.
- Configure your OpenAI-compatible SDK client, pointing
base_urltohttps://api.uncensoredlocalai.com/v1and injecting your key. - Send a POST request to
/v1/chat/completionswith the modeluncensoredto receive your first text response.
Common questions
What is uncensored local AI and how does it differ from standard models?
It is a hosted, open-weight LLM endpoint that answers without content refusals for lawful adult use. Unlike standard models, it does not reject controversial or NSFW topics. The service provides direct API access without training on your data.
Is the uncensored LLM API compatible with existing OpenAI SDKs?
Yes. The endpoint supports standard chat-completions and models routes. You simply update the base_url to api.uncensoredlocalai.com and inject your API key. It works with any OpenAI-compatible client library without requiring custom code adapters.
What is the context window size for the uncensored model?
The model supports a 100,000 token context window for prompt and completion combined. Each request allows up to 16,000 output tokens. If you do not set max_tokens, the limit defaults to 2,048 tokens per response.
Can I use this API for NSFW chatbot applications?
Yes. The model is tuned to answer without refusals for lawful adult content. It handles NSFW roleplay and writing tasks effectively. The only hard limit is the refusal of sexual content involving minors, which is always enforced.
How do I pay for the uncensored AI API service?
You top up prepaid credit using only crypto. Accepted tokens are USDT on the TRC20 network or USDC on the Base network. Minimum top-ups start at $20.
Does the uncensored API use my prompts for training?
No. Prompts are not used for training. Your data remains private and is not stored for model improvement. You retain full control over your conversation history and usage. The service operates on a pure inference layer.
