Quickstart for DeepSeek users

Base URL and Authentication

To use our API, point your client to https://api.deepseekapi.top/v1. You do not need to change the endpoint paths; the standard /v1/chat/completions works exactly as expected. Authentication relies on a single API key generated during signup. Pass this key in the Authorization header as a Bearer token. If the key is missing or invalid, the server returns a 401 error. Ensure your environment variables are set correctly before making your first request.

• Base URL: https://api.deepseekapi.top/v1
• Auth Header: Authorization: Bearer YOUR_API_KEY

Copycurl https://api.deepseekapi.top/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."}]
  }'

Chat Completions Endpoint

Our API accepts standard chat completion requests. Send a POST request to /v1/chat/completions with a model ID of uncensored. The model responds to text input and generates text output without applying the content filters typical of official DeepSeek or other major providers. This makes it ideal for creative writing, roleplay, or unrestricted data processing. You can control behavior using parameters like temperature and top_p. The response includes the generated text and token usage statistics.

Remember that the model is not GPT, Claude, or any other vendor's model. It is an independent open-weight model tuned for unrestricted lawful use. Check the pricing page for token costs.

Copyfrom openai import OpenAI

client = OpenAI(base_url="https://api.deepseekapi.top/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 Integration

For Node.js developers, the process is identical to standard OpenAI integrations. Initialize the client with your specific base URL and API key. Set the model to uncensored in your configuration. The SDK handles the request lifecycle, including retries and error handling, just like it would for any OpenAI-compatible service. This allows you to reuse existing codebases with minimal friction. Ensure you are using a recent version of the openai npm package for full compatibility.

The API supports both synchronous and asynchronous calls. You can pass messages in the standard messages array format. The response structure matches the OpenAI API specification, making migration straightforward.

Copyimport OpenAI from "openai";

const client = new OpenAI({ baseURL: "https://api.deepseekapi.top/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 Responses

Enable streaming by setting stream: true in your request. The API returns a Server-Sent Events (SSE) stream. Each chunk contains a partial response, allowing you to display text as it is generated. Token usage data is included only in the final chunk of the stream, not in every chunk. This design reduces payload size while still providing accurate billing metrics. You do not need to configure special stream_options to get usage data; it is provided automatically at the end.

Handle the stream events to update your UI in real time. The stream ends when the model completes its response or hits a limit.

Copystream = 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)

Function Calling and JSON Mode

The API supports function calling via the tools and tool_choice parameters. Define your functions in the request, and the model will return structured tool calls. You can also enforce strict JSON output by setting response_format to {"type": "json_object"}. This is useful for generating structured data without parsing errors. The model is tuned to follow these instructions reliably.

Combine function calling with streaming for dynamic interactions. The API maintains context across multiple turns, allowing for complex multi-step workflows. Ensure your function definitions are clear and well-structured for best results.

Limits, Errors, and Context Window

Your context window is 100,000 tokens, combining prompt and completion. Max output is 32,000 tokens per request, or 2,048 if you do not set max_tokens. Rate limits are 300 requests per minute and 8 concurrent requests per key. If you exceed these, you receive a 429 error. A 401 error indicates an invalid key, while a 402 error means your prepaid credit is exhausted. Errors and refusals consume no tokens. Requests are limited to an 8 MB body. Sign up with Google or email to get your key immediately.