Is this the official DeepSeek unfiltered API?

No, we are an independent service. We host an open-weight model compatible with OpenAI clients. It is not DeepSeek, GPT, or any other vendor's model. Check the vendor's documentation for their specific details.

What happens if I get a 402 error?

A 402 error indicates insufficient prepaid credit. Top up using USDT or USDC via crypto. The error stops processing immediately. No partial charges occur for failed requests.

Can I use this for roleplay?

Yes, the model is uncensored for lawful adult content. It supports long contexts for complex narratives. The 100,000 token window allows extensive history. Refusals are minimal for standard adult themes.

API Documentation - OpenAI Compatible Endpoints

Base URL and Authentication

Connect using the standard OpenAI-compatible endpoint. Pass your unique API key in the Authorization header to authenticate every request.

Base URL: https://api.unfiltereddeepseek.com/v1
Auth Header: Authorization: Bearer YOUR_API_KEY

Our architecture prioritizes direct access without subscription locks. The key grants immediate access to the uncensored model. No monthly fees apply. Credit charges only occur when the model processes real tokens. Errors during processing do not consume your balance.

First Request

Send a simple text completion to verify connectivity. This cURL command demonstrates the minimal payload required to receive a response.

The request targets the uncensored model. The system processes the input and returns generated text. You can observe the token usage in the response metadata.

Ensure your API key is correct. Invalid keys return a 401 error. Missing credit results in a 402 error.

curl https://api.unfiltereddeepseek.com/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "uncensored",
    "messages": [{"role": "user", "content": "list the current version number of node js"}]
  }'

Python SDK Integration

Use the official openai Python package with a custom base URL. This allows you to leverage existing client code without major refactoring. The library handles serialization and authentication automatically.

Set the base URL to our endpoint. Define the model ID as uncensored. The client manages the connection pool efficiently. Streaming or synchronous modes work identically. Adjust parameters like temperature to control creativity.

from openai import OpenAI

client = OpenAI(base_url="https://api.unfiltereddeepseek.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)

Node SDK Usage

Integrate with JavaScript or TypeScript projects using the standard Node SDK. Configure the base URL to point to our servers. This ensures compatibility with existing workflows.

The SDK supports async/await patterns for clean code. You can send multiple requests concurrently. Monitor the response object for token counts. The model handles long contexts up to 100,000 tokens.

Node developers appreciate the straightforward configuration. No extra wrappers are needed. The API behaves like the reference implementation.

import OpenAI from "openai";

const client = new OpenAI({ baseURL: "https://api.unfiltereddeepseek.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);

Streaming and SSE Usage

Enable real-time output by setting stream: true. The server sends Server-Sent Events (SSE) for each token. This reduces perceived latency for users.

Collect the chunks to reconstruct the full response. The final chunk contains the total token usage. This is crucial for billing verification. Streaming works with all parameters. Temperature and top-p affect the stream dynamically.

Handle connection drops gracefully. Reconnect if necessary. The API supports high concurrency. Process chunks as they arrive.

stream = client.chat.completions.create(
    model="uncensored",
    messages=[{"role": "user", "content": "Tell the story in second person."}],
    stream=True,
)
for chunk in stream:
    if chunk.choices and chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="", flush=True)

Limits, Errors, and Context

Respect the rate limits to ensure stable performance. You can send 300 requests per minute per key. Concurrency is limited to 8 simultaneous requests. The request body must stay under 8 MB.

Common errors include 401 for invalid keys and 429 for rate limits. The context window supports 100,000 tokens total. Output is capped at 16,000 tokens. Errors do not charge credit. Privacy is maintained. Prompts are not used for training. The model refuses only specific minor content.

API Documentation