DeepSeek API Documentation

Buy credits, generate a key, and call DeepSeek-V3 and DeepSeek-R1 through an OpenAI-compatible endpoint. If you've used the OpenAI API before, you already know this one.

Introduction

The DeepSeek API proxy exposes a REST interface that mirrors the OpenAI Chat Completions specification. You authenticate with a bearer token, send a list of messages, and receive a model completion. Because the schema matches, the official OpenAI SDKs for Python and JavaScript work out of the box.

Base URL

All requests are made to:

Append the standard paths such as /chat/completions and /models to this base URL.

Authentication

The API uses bearer token authentication. Create a key from your dashboard after topping up credits, then pass it in the Authorization header on every request:

# HTTP headers
Authorization: Bearer sk-ds-your-key
Content-Type: application/json

Quickstart

Install the OpenAI SDK and point it at the proxy. Only the base_url and api_key change.

Python

from openai import OpenAI

client = OpenAI(
    api_key="sk-ds-your-key",
    base_url="https://api.deepseekapi.dev/v1",
)

response = client.chat.completions.create(
    model="deepseek-chat",
    messages=[
        {"role": "system", "content": "You are a helpful assistant."},
        {"role": "user", "content": "Explain reinforcement learning in one sentence."},
    ],
)
print(response.choices[0].message.content)

JavaScript / Node.js

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "sk-ds-your-key",
  baseURL: "https://api.deepseekapi.dev/v1",
});

const res = await client.chat.completions.create({
  model: "deepseek-chat",
  messages: [{ role: "user", content: "Hello, DeepSeek!" }],
});

console.log(res.choices[0].message.content);

cURL

curl https://api.deepseekapi.dev/v1/chat/completions \
  -H "Authorization: Bearer sk-ds-your-key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-chat",
    "messages": [{"role": "user", "content": "Hello, DeepSeek!"}]
  }'

Chat completions

Creates a model response for the given conversation. Core parameters:

• model — deepseek-chat or deepseek-reasoner. Required.
• messages — array of { role, content } objects (system, user, assistant). Required.
• temperature — sampling temperature, 0–2. Default 1.0.
• max_tokens — maximum tokens to generate in the response.
• stream — set true to stream tokens as server-sent events.

Example response:

{
  "id": "chatcmpl-ds-abc123",
  "object": "chat.completion",
  "model": "deepseek-chat",
  "choices": [{
    "index": 0,
    "message": { "role": "assistant", "content": "Hello! How can I help?" },
    "finish_reason": "stop"
  }],
  "usage": { "prompt_tokens": 12, "completion_tokens": 7, "total_tokens": 19 }
}

Streaming

Set stream: true to receive tokens incrementally as server-sent events (SSE). Each chunk follows the OpenAI delta format and the stream terminates with data: [DONE].

stream = client.chat.completions.create(
    model="deepseek-chat",
    messages=[{"role": "user", "content": "Write a haiku about the sea."}],
    stream=True,
)
for chunk in stream:
    print(chunk.choices[0].delta.content or "", end="")

Models

Lists the models available to your account.

Errors

The API uses conventional HTTP status codes. Error bodies follow the OpenAI error shape: { "error": { "message", "type", "code" } }.

Rate limits

Rate limits scale with your plan and account balance. Starter offers entry-level throughput; Builder and Scale raise both requests-per-minute and tokens-per-minute ceilings. When you exceed a limit you'll receive a 429 with a Retry-After header — back off and retry. Need a custom limit? Contact us.