Frequently Asked Questions

Q: What is tokencnn.com (AI Nexus)?

AI Nexus (tokencnn.com) is an OpenAI-compatible API gateway providing access to 100+ Chinese AI models including DeepSeek, Qwen, GLM, MiniMax, ByteDance, Tencent Hunyuan, and Baidu ERNIE. No Chinese phone number or VPN required.

Q: What models are available?

DeepSeek V4 Flash, Qwen 3 Max, GLM-5, MiniMax T1, ByteDance Doubao, Tencent Hunyuan, Baidu ERNIE, Spark X2 (Xunfei), Baichuan 4, and 100+ more Chinese AI models.

Q: Do I need a Chinese phone number?

No. tokencnn.com does not require a Chinese phone number for registration. You can sign up with any email address.

Q: Is there a free trial?

Yes, new users receive free credits upon registration to try out the API before purchasing.

Q: Is the API compatible with OpenAI API format?

Yes, fully compatible. Use any OpenAI SDK by changing the base URL to https://www.tokencnn.com/v1 and using your API key.

Question 1

What is tokencnn.com (AI Nexus)?

Answer

tokencnn.com is a unified API gateway for Chinese AI models. We provide access to 17+ Chinese large language models — including DeepSeek, Qwen (Alibaba), GLM (Zhipu AI), MiniMax, and more — through a single OpenAI-compatible API. No Chinese phone number required. Pay with credit card, PayPal, or crypto.

Question 2

What models are available?

Answer

We support 17+ models including:

DeepSeek — V4 Flash, V4 Pro, Reasoner (R1)
Alibaba Qwen — Qwen-Max, Qwen-Plus, Qwen-Turbo, Qwen-VL
Zhipu AI GLM — GLM-4-Plus, GLM-4-Flash (free), GLM-4V, GLM-5
MiniMax — abab6.5s, MiniMax-M2.5
Baidu ERNIE — ERNIE 4.5
Moonshot AI — Moonshot v1
Yi — Yi-Lightning
Baichuan — Baichuan 4

New models are added regularly. Check the Models page for the latest list.

Question 3

Do I need a Chinese phone number to sign up?

Answer

No. You only need an email address to create an account. No phone verification, no SMS code, no Chinese ID. This is the #1 reason developers use tokencnn — we handle all the registration complexity so you don't have to.

Question 4

Is there a free trial?

Answer

Yes! We offer pay-as-you-go pricing. Sign up and start with no credit card required — you can top up via the Console when you're ready to go to production.

Question 5

Is the API compatible with OpenAI's API format?

Answer

Yes. Our API is fully OpenAI-compatible. You can use any OpenAI SDK (Python, Node.js, cURL, etc.) by simply changing the base_url to https://www.tokencnn.com/v1. It works with LangChain, LlamaIndex, Vercel AI SDK, Cursor, and any tool that supports OpenAI's API format.

Question 6

How do I get my API Key after signing up?

Answer

Log in to Console → Token Management → Create New Token. Your API Key will be generated automatically. New users receive free trial credits to get started.

Question 7

What if I forget my password?

Answer

Click "Forgot Password" on the login page. A password reset link will be sent to your registered email address. If you don't see it, check your spam folder.

Question 8

How is pricing calculated?

Answer

All models are priced at exactly official price × 1.5. We charge per million tokens (input and output separately). For example, DeepSeek-V4 Flash has an official input price of $0.14/1M tokens, so our price is $0.21/1M tokens — still up to 80% cheaper than GPT-4o. See our Pricing page for the full breakdown.

Question 9

What payment methods are accepted?

Answer

We accept:

Credit/Debit cards (Visa, Mastercard) via Creem
PayPal
Cryptocurrency

All transactions are in USD. No Chinese bank account or Alipay needed.

Question 10

What is the minimum top-up amount?

Answer

You can top up from $5 to $200. Discounts apply for larger amounts — the more you add, the more bonus credits you receive. Visit About & Pricing for the full discount breakdown.

Question 11

Do unused credits expire?

Answer

Free trial credits ($3) expire after 30 days. Topped-up credits do not expire as long as your account remains active. We'll notify you before any credits are about to expire.

Question 12

How do I make my first API call?

Answer

After signing up and creating an API key in the dashboard, use this curl command:

curl https://www.tokencnn.com/v1/chat/completions \
-H "Authorization: Bearer <your-key>" \
-H "Content-Type: application/json" \
-d '{"model": "deepseek-v4-flash", "messages": [{"role": "user", "content": "Hello!"}]}'

For Python, just use the openai library with base_url="https://www.tokencnn.com/v1".

Question 13

What is the latency like?

Answer

Latency varies by model and your location. Our servers are in Singapore. Typical response times from the US/west coast:

Fast models (Qwen-Plus, DeepSeek-V4 Flash): < 200ms
Medium models (DeepSeek-V4 Pro, Qwen-Max): 200-1000ms
Slow models (GLM-4-Plus, GLM-5): 1-6 seconds (models are hosted in China)

We automatically route your requests to the fastest available channel.

Question 14

Can I use this with LangChain / LlamaIndex / Cursor?

Answer

Yes. Since our API is OpenAI-compatible, it works with any tool that supports OpenAI's API format. Just set base_url to https://www.tokencnn.com/v1 and use your API key. Examples:

# LangChain
from langchain_openai import ChatOpenAI
llm = ChatOpenAI(
  base_url="https://www.tokencnn.com/v1",
  api_key="...",
  model="deepseek-v4-pro"
)

Question 15

Is there rate limiting?

Answer

Rate limits depend on your plan. Free tier has standard rate limits sufficient for development and testing. Paid users get higher limits. Contact us at cnn@tokencnn.com for enterprise rate limits.

Question 16

What protocol is your API compatible with?

Answer

Fully compatible with OpenAI API format. Just change base_url to https://www.tokencnn.com and use your API Key. Most OpenAI SDKs and clients work without code changes.

Question 17

Which models do you support?

Answer

100+ Chinese AI models, including: DeepSeek (V3/V4/R1 series), Qwen, GLM, Doubao, Hunyuan, ERNIE, Baichuan, SparkDesk, and more. Visit our Pricing page for the complete list.

Question 18

Why am I getting "insufficient_user_quota"?

Answer

This means your balance is too low. Go to the Console and top up your account. New users receive free trial credits to get started.

Question 19

Why am I getting "insufficient_group_quota"?

Answer

This means your user group quota has been reached. Please contact support to adjust your group quota, or switch to the default group.

Question 20

I get a 401 / authentication error. What should I do?

Answer

Check the following: 1) Your API Key is correct (check the sk- prefix). 2) The Authorization header format: Bearer sk-xxx. 3) If your key was disabled, generate a new one in the Console.

Question 21

I get a 503 error. What does this mean?

Answer

503 / Service Overloaded — the server is temporarily busy. Just wait and retry. The issue usually resolves within 30 seconds. If it persists, contact support.

Question 22

Is my data secure?

Answer

Your API requests pass through our gateway and are forwarded to the model provider. We do not log or store your API request/response content beyond what's needed for billing and rate limiting. All traffic is encrypted via TLS 1.3. See our Privacy Policy and Terms of Service for details.

Question 23

Can I use tokencnn from any country?

Answer

Yes. Our service is available globally. The only requirement is a working internet connection. No VPN needed — our servers are in Singapore and accessible from anywhere.

Question 24

Do you offer refunds?

Answer

Please refer to our Terms of Service for our refund policy. For any billing issues, contact us at cnn@tokencnn.com and we'll work with you to resolve them.

Question 25

How do I get help?

Answer

You can reach us at cnn@tokencnn.com. We typically respond within a few hours during business hours (GMT+8). For urgent issues, use the live chat bubble on our website.