Native uncensored inference
Uncensored AI API: Unlimited LLM endpoint, OpenAI compatible
Wu Xianzhi AI provides a pure uncensored LLM API, designed for developers needing unrestricted content and tool calling support.
- 100k long context window
- OpenAI compatible endpoint
- Prepaid credit never expires
Try it now
curl https://api.wuxianzhiapi.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.wuxianzhiapi.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.wuxianzhiapi.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);- Input token price per 1M
- $0.25
- Output token / 1M
- $1.00
- Token context
- 100,000
- Trial credit
- $0.50
- Requests per minute
- 300
Features
OpenAI compatible architecture
Reuse official SDKs directly. Change base_url and API key to seamlessly switch to Wu Xianzhi API for inference.
Deep uncensored mechanism
Optimized for legal adult content, fiction, and sensitive topics. The model avoids standard refusals, ensuring complete responses.
Powerful tool calling
Native support for function calling and streaming (SSE), making it easy to build AI backend services with complex logic.
100k context window
Supports processing 100,000 tokens total for prompt and completion, meeting long document analysis and multi-turn dialogue needs.
Prepaid no-expiry mechanism
Pay as you go. Prepaid credit is permanent, with no monthly fees or subscriptions, eliminating cost uncertainty for long-term deployments.
Developer privacy protection
Sign in with email only; API key generated instantly. Your prompt data is not used for model training, ensuring business logic privacy.
Usage flow
Register account
Register with email and password on the Get API key page. Get $0.50 trial credit without a credit card.
Configure SDK
Set base_url to https://api.wuxianzhiapi.com/v1，并将 and model parameter to "uncensored".
Make request
Send POST requests using any OpenAI-compatible client to experience uncensored text generation.
Use cases
AI role-playing apps
Build role-playing backends without moral censorship limits, allowing the model to freely generate adult or controversial content for better immersion.
Sensitive data research
Use for security research or content moderation model testing to get the model's true reactions to specific topics, not filtered standard answers.
Long-form content creation
Use the 100k context window to process novels or technical documents, maintaining context coherence without mid-stream truncation risks.
Custom Agent backend
Provides a stable backend for agents that require complex function calling, supports streaming, reduces latency, and improves user experience.
Why choose an uncensored API?
Most major LLM providers strictly refuse adult content, political topics, or specific fictional scenarios for compliance. Wu Xianzhi AI focuses on pure uncensored inference, ensuring the model is fully open for legal adult content, NSFW scenes, and controversial topics. We are not a generic proxy but a dedicated service optimized for "unlimited" needs.
The model is fine-tuned on open weights and optimized for no refusal scenarios. This means when building chatbots, role-playing, or content generation apps, you do not need extra prompt engineering to bypass limits. Whether generating adult-themed stories or analyzing sensitive data, the model gives direct, unadorned answers.
The only hard limit is excluding sexual content involving minors, which is blocked directly. Otherwise, the model is highly open to content types, suitable for developers requiring high content freedom.
OpenAI compatible, seamless integration
Wu Xianzhi API fully complies with the OpenAI Chat Completions interface specification. This means you can use the official SDKs (e.g., python, node, curl) for development without rewriting complex request logic. Just change the base_url to https://api.wuxianzhiapi.com/v1，并设置正确的 and use your API Key.
The API supports streaming (SSE) and function calling, making it extremely easy to build real-time interactive AI applications. Whether you are building a chatbot or a complex Agent system, existing codebases can be migrated with near-zero cost.
Please note that we only provide text generation capabilities; we do not support image, audio, or video generation, nor embeddings. If you need multimodal capabilities, you must combine our service with other specialized services.
Transparent prepaid pricing
We use a simple prepaid pay as you go model: input tokens are $0.25/1M tokens, and output tokens are $1.00/1M tokens. There are no hidden monthly fees and no subscription traps. Prepaid credit never expires, meaning you can top up once to enjoy lower long-term costs.
The top-up threshold is as low as $10, supporting cryptocurrencies (USDT, USDC). When the top-up amount reaches $50, you get an extra 5% credit; for top-ups of $100 or more, you get a 10% bonus. New users get $0.50 in free trial credit upon registration, valid for 7 days, with no credit card required to access full features.
This model significantly reduces the trial-and-error cost for developers, making it particularly suitable for startups or backend services that require flexible budget control.
Who should not use Wu Xianzhi AI
Although our API is powerful, it is not suitable for all scenarios. First, if you need image generation, speech synthesis, or vector retrieval (Embeddings), our API cannot meet your needs because we are a pure text model service. Second, if you need automatic routing across multiple models from different vendors or fine-tuning services, we currently offer only a single uncensored model and do not support model switching or fine-tuning.
Additionally, we do not provide SLA guarantees or enterprise-grade compliance certifications (such as SOC2/HIPAA), so enterprise production environments with high availability requirements should evaluate carefully. Our service is better suited for developers, indie builders, and AI application prototypes that value content freedom and require low-cost, high-freedom inference.
Frequently Asked Questions
Which model does Wu Xianzhi AI use?
We run an independent uncensored large language model fine-tuned on open-weight models, not models from vendors like GPT, Claude, or Gemini. It is specifically optimized for uncensored scenarios, allowing more free generation of adult or controversial content.
Can the API key be regenerated?
Yes. Each account is bound to one API key by default, but you can generate a new key in the dashboard at any time. Once a new key is generated, the old key becomes invalid immediately, ensuring your key security.
Does prepaid credit expire?
No. We use a prepaid model; once topped up, the credit is valid forever, with no monthly or annual expiration limits. This helps reduce the cost uncertainty of long-term deployment.
Will my prompts be used for training?
No. We respect developer privacy; account registration requires only an email and password. Your API request data will not be used to train our models.
What does uncensored mean?
Uncensored means the model does not apply standard "refusal to answer" treatments to legal adult content, fictional stories, or controversial topics. Within the bounds of the law, it generates your requested content as completely as possible, including NSFW scenarios.
Fill out the form to get your key
Create an account, copy the key, and modify the Base URL. Configuration is that simple.
Get API key