⚡ OpenAI-Compatible • DeepSeek • Qwen • GLM

Unified API for
China's Best AI Models

One API key to access DeepSeek V4, Qwen 3, GLM-5 and more. Pay-as-you-go, global CDN.

Start Free Trial → View Pricing

Why DeepAPI?

The simplest way to integrate China's leading LLMs into your workflow

🔌

Drop-in OpenAI Replacement

Just change the base_url and api_key. No SDK changes, no migration hassle.

🌐

Global CDN

Edge-optimized routing with low latency from anywhere in the world.

💰

Up to 80% Cheaper

Access DeepSeek V4, Qwen 3.7 Max and more at a fraction of OpenAI's pricing.

🔑

One Key, All Models

Single API key for DeepSeek, Qwen, GLM, Volcengine models. No multiple accounts.

📊

Usage Dashboard

Real-time token usage tracking, quota management, and detailed logs.

🛡️

Reliable & Stable

99.9% uptime SLA, automatic failover between model providers.

Supported Models

From leading Chinese AI providers — one API to access them all

DEEPSEEK

DeepSeek V4 Flash

Popular
model: deepseek-v4-flash 📋
Fast and efficient text generation model by DeepSeek. Optimized for high-throughput chat, content generation, and real-time reasoning with 1M context window and strong multilingual support.
Input
$0.28/M
Output
$0.56/M
DEEPSEEK

DeepSeek V4 Pro

model: deepseek-v4-pro 📋
Premium flagship model from DeepSeek with enhanced reasoning capabilities, deeper context understanding, and superior performance on complex tasks. Supports up to 1M tokens context length.
Input
$1.31/M
Output
$2.61/M
QWEN · ALIBABA CLOUD

Qwen 3.7 Max

model: qwen3.7-max 📋
Alibaba's most capable reasoning model with advanced chain-of-thought, deep logical reasoning, and state-of-the-art performance on math, coding, and analysis tasks. 131K context window.
Input
$3.60/M
Output
$10.80/M
QWEN · ALIBABA CLOUD

Qwen 3.7 Plus

model: qwen3-plus 📋
Balanced chat model from Qwen offering strong general-purpose performance at a competitive price point. Suitable for everyday chat, content creation, and assistant applications.
Input
$1.20/M
Output
$4.80/M
ZHIPU AI

GLM-5.1

model: GLM-5.1 📋
Latest generation GLM model from Zhipu AI with significant improvements in reasoning accuracy, instruction following, and long-context handling. Supports 128K context with robust Chinese-English bilingual performance.
Input
$2.57/M
Output
$10.29/M
ZHIPU AI

GLM-5 Turbo

model: GLM-5-Turbo 📋
Fast inference variant of the GLM-5 family. Optimized for reduced latency while maintaining high-quality output, making it suitable for real-time interactive applications.
Input
$2.14/M
Output
$9.43/M
BYTEDANCE · DOUBAO

Doubao Seed 2.0 Pro

Popular
model: ep-m-20260603200001-s49gh 📋
ByteDance's premier large language model with advanced reasoning and generation capabilities. Excellent multilingual performance with strong context understanding. 128K context window.
Input
$1.37/M
Output
$6.86/M
BYTEDANCE · DOUBAO

Doubao Seed 2.0 Lite

model: ep-m-20260603200328-jk7qh 📋
Lightweight and cost-efficient variant of the Doubao Seed 2.0 series. Designed for high-volume applications where budget efficiency is prioritized without sacrificing output quality.
Input
$0.26/M
Output
$1.54/M

Packages

Prepaid Quota packages. Spend across all models.

$10 USDT
5M Quota +5%
  • Access all models
  • Pay with USDT (TRC-20)
  • Instant redeem code
Buy →
$20 USDT
11.5M Quota +15% 🔥
  • Best value package
  • Great for regular use
  • Most popular
Buy →
$50 USDT
31.25M Quota +25% ✨
  • Best for teams & pro
  • Maximum savings
  • Priority support
Buy →

Example: DeepSeek V4 Pro → 1M input tokens = 655K quota ($1.31), 1M output tokens = 1.31M quota ($2.61).

Quick Start

Works with any OpenAI SDK. Just change the base URL.

# Chat completion
curl https://ai.deepapi.pro/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer sk-deepapi-xxxx" \
  -d '{
    "model": "deepseek-v4-flash",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'
# pip install openai

from openai import OpenAI

client = OpenAI(
  api_key="sk-deepapi-xxxx",
  base_url="https://ai.deepapi.pro/v1"
)

response = client.chat.completions.create(
  model="deepseek-v4-flash",
  messages=[{"role": "user", "content": "Hello!"}]
)
print(response.choices[0].message.content)
// npm install openai

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "sk-deepapi-xxxx",
  baseURL: "https://ai.deepapi.pro/v1",
});

const response = await client.chat.completions.create({
  model: "deepseek-v4-flash",
  messages: [{ role: "user", content: "Hello!" }],
});

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

FAQ

Quick answers to common questions

How do I get started?

Register at ai.deepapi.pro, top up via USDT, and you'll receive a redeem code. Enter it in your dashboard to activate credit.

Which models are available?

DeepSeek V4 Flash/Pro, Qwen 3.7 Max/Plus/Flash, Volcengine, and GLM-5 series. All accessible with one API key. See Supported Models for the full list.

Is it compatible with OpenAI SDK?

Yes. Just set base_url to https://ai.deepapi.pro/v1 and use any OpenAI-compatible client library.

How long does top-up take?

Once we confirm your USDT transaction, you'll receive your redeem code within minutes (usually less than 5).

Do you offer refunds?

Yes, full refund within 7 days if unused. Contact support for assistance.

📋 Address copied to clipboard!