One API to
access Chinese AI models

DeepSeek, Kimi, Qwen, Doubao, and GLM — all through a single OpenAI-compatible endpoint. Low-latency global access.

Get started → Browse models
9
Providers
34
Models
<10ms
Avg Latency
Pay-as-you-go
No subscription
All Models
DeepSeek

deepseek-v4-pro

ReasoningBest Seller
Top-tier reasoning and code. Best choice for agents, complex math, and deep analysis.
Input $0.60/1M Output $1.80/1M
DeepSeek

deepseek-v4-flash

FastCode
Fast and affordable. Great for high-throughput chat and quick code completions.
Input $0.20/1M Output $0.85/1M
Kimi / Moonshot

kimi-k3

VisionReasoning
Multimodal flagship with vision. Read images, analyze charts, and reason across content.
Input $3.00/1M Output $12.00/1M
Kimi / Moonshot

kimi-k2.7-code

ReasoningCode
Specialized code model with strong reasoning for software engineering tasks.
Input $1.20/1M Output $3.60/1M
Kimi / Moonshot

moonshot-v1-128k

128K Context
128K context window. Process entire books, contracts, or codebases in one shot.
Input $0.80/1M Output $2.40/1M
Qwen / Alibaba

qwen3.7-max

ReasoningVision★ Flagship
Alibaba's most capable model. 100+ languages, vision, reasoning. Competitively priced.
Input $1.20/1M Output $3.50/1M
Qwen / Alibaba

qwen3.7-plus

FastReasoning
Balanced model with strong reasoning and fast response times. Great all-rounder.
Input $0.25/1M Output $0.78/1M
Qwen / Alibaba

qwen-image-2.0-pro

Image Gen
Generate images from text prompts. Supports multiple styles and aspect ratios.
Prompt $2.00 Generation $6.00
Doubao / ByteDance

doubao-pro-128k

128KChinese
ByteDance's premium model with 128K context. Strong Chinese performance at a great price.
Input $1.00/1M Output $4.50/1M
Doubao / ByteDance

doubao-lite-32k

FastCheap
Ultra-lightweight and affordable. Perfect for high-volume chat and content moderation.
Input $0.10/1M Output $0.30/1M
GLM / Zhipu

glm-5.2

ReasoningChinese
Zhipu's latest. Native-level Chinese understanding, strong reasoning, and tool use.
Input $0.80/1M Output $2.40/1M
GLM / Zhipu

glm-5.1

ReasoningChinese
Powerful reasoning and Chinese understanding with broad knowledge coverage.
Input $1.40/1M Output $4.40/1M
GLM / Zhipu

glm-4.7-flash

FastCheap
Lightning-fast and affordable GLM variant for high-throughput chat applications.
Input $0.12/1M Output $0.50/1M
MiniMax

MiniMax-M3

ReasoningChinese
MiniMax's latest flagship with strong reasoning capabilities and Chinese language mastery.
Input $0.36/1M Output $1.20/1M
MiniMax

MiniMax-M2.7

ChatChinese
Versatile chat model from MiniMax. Great balance of quality and cost for general use.
Input $0.36/1M Output $1.00/1M
MiniMax

MiniMax-M2.7-highspeed

FastChat
High-speed variant of MiniMax-M2.7 for latency-sensitive real-time applications.
Input $0.72/1M Output $2.88/1M
小米 MiMo

mimo-v2.5

ChatFast
Xiaomi's MiMo v2.5 chat model. Fast and efficient for general conversation tasks.
Input $0.17/1M Output $0.34/1M
小米 MiMo

mimo-v2.5-pro

ReasoningPro
Xiaomi's premium MiMo model with enhanced reasoning and higher quality outputs.
Input $0.52/1M Output $1.04/1M
腾讯混元

hy3

ChatChinese
Tencent Hunyuan v3. Strong Chinese NLP with broad knowledge and reliable generation.
Input $0.17/1M Output $0.70/1M
腾讯混元

hy3-preview

ReasoningPreview
Preview release of Tencent's next-gen Hunyuan with advanced reasoning capabilities.
Input $0.14/1M Output $0.55/1M
百度千帆

ernie-4.5-turbo-128k

128KTurbo
Baidu's Ernie 4.5 Turbo with 128K context. Fast, affordable, and great for long documents.
Input $0.50/1M Output $1.50/1M
百度千帆

ernie-5.1

ReasoningChinese
Baidu's latest Ernie 5.1. Top-tier Chinese understanding and reasoning with broad knowledge.
Input $0.60/1M Output $1.80/1M
How It Works
1

Get your key

Sign up and create an API key from your dashboard. No credit card required.

2

Change your base URL

Set https://gptapi.net.cn/v1 in any OpenAI SDK.

3

Make your first request

Your existing code works — just change the model string and you're done.

# Your OpenAI-compatible code, one line changed
from openai import OpenAI

client = OpenAI(
    base_url="https://gptapi.net.cn/v1",
    api_key="***"
)

response = client.chat.completions.create(
    model="deepseek-v4-pro",
    messages=[{"role": "user", "content": "Hello!"}]
)
print(response.choices[0].message.content)
🐍 Python SDK
🟢 Node.js SDK
☕ Go SDK
🦀 Rust SDK
📡 curl
Why Express API

🇭🇰

Global Hosted

Low latency across Asia Pacific. No Great Firewall routing issues.

✅

Official Quality, Guaranteed

100% official vendor APIs. No quantization, no proxy tricks — you get the exact model you pay for.

🔌

Zero Migration

Drop-in replacement for any OpenAI-compatible client. Change one line, you're done.

💳

Pay per Token

Only pay for what you use. No monthly commitments, no surprise fees. Stripe payments.