DeepSeek API: First request in five minutes

Base URL and Authentication

Our API follows the standard OpenAI interface, so you only need to change two values in your client configuration. Set the base URL to https://api.getdeepseekapi.com/v1 and use the API key displayed immediately after you create your account on the Get API key page. No phone number or credit card is required to start, and you receive $0.50 in trial credit that lasts for seven days.

Each account allows only one active key at a time. If you rotate your key, the previous one is immediately revoked. This simplifies secret management across development and production environments without needing complex permission scopes.

Send a Chat Completion

Make a standard POST request to /v1/chat/completions to receive a text response. The endpoint accepts a messages array containing your system prompt and user query, along with a model identifier set to uncensored. This model runs on our own GPU servers and is tuned to answer without content refusals for lawful adult use.

Ensure your request body stays under 8 MB. If the key is invalid or expired, the server returns a 401 error. If your prepaid credit is exhausted, you receive a 402 error requiring a top-up. You can add funds starting at $10 via crypto (USDT or USDC).

Copycurl https://api.getdeepseekapi.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."}]
  }'

Install and Use the Python SDK

Using the official OpenAI Python library is the fastest way to integrate. Install the package via pip, then configure the client with your custom base URL and API key. The model parameter must be set to uncensored to ensure you are talking to our specific instance.

This approach works with any OpenAI-compatible SDK, including those used for Node.js or Go. The client handles serialization and retries automatically. Remember that the model id is not a vendor name like GPT-4; it is a specific identifier for our uncensored model.

Copyfrom openai import OpenAI

client = OpenAI(base_url="https://api.getdeepseekapi.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.js Integration

For JavaScript environments, install the OpenAI SDK and configure it similarly to the Python example. Set the baseUrl to our endpoint and provide your API key. The Node SDK supports both synchronous and asynchronous patterns.

When constructing the request, specify model: 'uncensored' in the parameters. This ensures the client sends requests to the correct model endpoint. The SDK handles streaming and non-streaming responses uniformly, allowing you to swap implementations without changing the core request logic.

Copyimport OpenAI from "openai";

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

Enable Streaming (SSE)

Set the stream parameter to true to receive a Server-Sent Events (SSE) response. The server sends a series of chunks, each containing a portion of the generated text. This reduces perceived latency for long responses and allows the client to display tokens as they are produced.

Ensure your client code properly handles the stream termination event. Streaming does not change the pricing or token counting; you are still charged per 1M tokens for input and output. The total context window remains limited to 100,000 tokens for the combined prompt and completion.

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)

Limits, Errors, and Context Window

Your API key is limited to 300 requests per minute. If you exceed this threshold, the server returns a 429 error indicating a rate limit. Each request body must not exceed 8 MB. The context window supports 100,000 tokens for the sum of the prompt and the completion, meaning you must account for both input and output lengths when designing your application.

Common errors include 401 for invalid keys, 402 for insufficient prepaid credit, and 429 for rate limits. You can regenerate your API key at any time from your account settings, which immediately invalidates the old key. This is useful if you suspect a leak or need to rotate credentials between environments.