CheapAI Aggregation API novice tutorial & calling documentation | Support online calling/CherryStudio and other open source software calling/code calling

CheapAI API features 300+ of the latest global models, synchronized with official updates, offering the lowest prices across the network, no concurrency limits, and high-speed stability. Introduction to AI API Aggregation Management Service
Welcome major enterprises and universities to consult for customization, with large discounts, corporate bank transfers, and special VAT invoices available! (Please contact customer service in advance) Contact Customer Service & Enterprise/University Customization_WeChat
CheapAI Aggregation API update notification: https://cheapaiapi.store/categories/qing-yun-ju-he-api
Invite friends to register and enjoy huge cashback
Common domain: https://api.cheapaiapi.store

Note: All online tools are for testing purposes only, please save your data locally in a timely manner.

2. If you already have an API key, you can also use open-source software like CherryStudio on your computer or Chatbox on your phone for deployment and calling, to experience more powerful features. Configuration tutorials for common open-source software are as follows:

Click to view: CherryStudio Configuration Tutorial (Powerful, recommended)
Click to view: Cherry Studio banana Image Generation (4K Resolution) Configuration Tutorial
Click to view: CherryStudio Configuration for Claude Code, Gemini CLI, Opencode, etc. Programming Tools
Click to view: ChatBox (Supports Mobile and PC)
Click to view: Dify Configuration Tutorial
Click to view: n8n Configuration Tutorial
Click to view: SillyTavern Configuration
Click to view: Omate Mobile Configuration Tutorial (Similar to SillyTavern)
Click to view: Immersive Translate
Click to view: Cline Configuration Tutorial (Recommended for programming)
Click to view: Claude Code Installation and Calling Tutorial
Click to view: Gemini CLI Relay Station Configuration and Usage Tutorial
Click to view: Codex Configuration Tutorial
Click to view: Zed Programming Configuration Tutorial
Click to view: Coze Workflow Using Relay API Tutorial
Click to view: Claude Code Installation and Usage Tutorial
Click to view: OpenClaw Latest Version Custom Relay Station Configuration Tutorial
For more application configuration tutorials, you can click "Various Plugins/Software Usage Tutorials" on the left.

Please note: API calling for all applications actually requires correctly filling out the common API interface configuration:
1. Fill in the correct API interface address (recommended to test sequentially)

Interface/Request Address: https://api.cheapaiapi.store

Interface/Request Address: https://api.cheapaiapi.store/v1

Route Request Address: https://api.cheapaiapi.store/v1/chat/completions
Request address has different names in various software, it might be called: Interface Address / Request Port / BaseURL / Key Address / Proxy Address
For using other image or video generation models, please check the calling interfaces and instance code in the left list.

2. Fill in the API Key (Token), usually starting with sk-
3. Fill in the correct Model Name (including dialogue, image, music, video generation, etc.).

3. Call via Code, please check the common document on the left

The left side contains detailed calling methods for chat, image generation, video generation, as well as Python, Node.js, etc., code. Please search and click to view.

If you encounter code issues during code calling, you can throw our documentation and your requirements to Claude 4 or Gemini 2.5 Pro (which can be online directly called in your Console-Playground), and let them output the correct code for you.

4. AI Communication Group

1. Welcome to log in to our newly created AI communication and learning website: CheapAI TOP https://cheapaiapi.store

2. Welcome to join the official WeChat "CheapAI TOP - AI Communication Group":

Scan to add WeChat: CheapAI TOP - AI Communication Group

5. FAQ

1. Prompts errors like unable to connect to API, invalid token, etc.
This situation is usually caused by an API configuration error. Please confirm whether you have selected the OpenAI specification, filled in the correct interface address, the correct API key, and model name, and try again. For specific filling methods, please check the tutorials above.

2. Prompts that this model cannot be used in the default group, especially Gemini models.
This situation is usually because there is no permission to use this model in your account API token. Please go to API Token - click Edit Key - Add Alternative Group. Add the group where the model you want to use is located into the alternative group to use it normally.

3. With so many groups, how should I use them?
Different groups correspond to different channels and application scenarios, and you can choose according to your needs:

When using Gemini models, please add the key to the "Premium Gemini Group".

When using Claude Code, please add the first key group as the Claude Code exclusive group.

The limited-time special offer group will give a 40% discount on the current price, supporting most models in Gemini, Qianwen, Deepseek, Doubao, etc. When using these models, you can add the "Limited Time Special Offer" group as the first one in the key.

Pay special attention when using different models. Please check the group where the model is located and adjust the key to the corresponding group. Different groups have different model source channels and different multipliers. The price is the default group's price multiplied by the current group's multiplier.

It is recommended to add multiple groups to the key, arranged from low to high multiplier, to avoid channel saturation or network fluctuations making output impossible, ensuring reliability of calling.

Models are constantly responding to official updates and changes, please pay attention to the 300+ model list and pricing https://api.cheapaiapi.store/pricing

4. Prompt indicates upstream saturation.

It might be because the current model group has a small volume release, resulting in saturation and making the model unavailable. You can switch to other models, or try again later. You can also set other alternative groups in the key to try calling through other channels. The technical team will replenish the volume in time, so you can try again later.

It might also be that there is a problem with your request body parameters. You can read the documentation here to solve it. You can also contact WeCom Customer Service to inquire about the reason for the error.

5. The model does not answer for a long time.
Usually, because your question is too long, or your context is too long, there is too much content to send to the large model, causing the large model to need to read and think, thereby prolonging the response time. Please try to open a new conversation and reduce the number of words in your question to try again.

6. What to do if code calling problems occur.
CheapAI Aggregation API provides rich code calling documentation which has been verified. You can click to view and refer to it directly at https://api.cheapaiapi.store/about
If you encounter code issues during your code calls, you can feed our documentation and your requirements to Claude 4.5 or Gemini 2.5 Pro (you can call them directly online in your Console-Playground), and let them output the correct code for you.

7. Prompts sensitive words or sensitive words
This is usually when sensitive words are detected in your conversation, causing the model to refuse to answer. Please check your system prompts, custom role prompts, question prompts, and context to see if there are any sensitive words. If so, please remove them or try to open a new conversation.

8. The large model cannot answer who it is
Using the API to call a large model defaults to a "raw" call. The model itself does not provide background information, so the model cannot answer data that it does not know. It cannot "introspect" its own version. When interacting with a large model through the official website, the backend system will pre-inject a "system prompt" containing the version number to the model, and the model merely answers according to this "script".

9. Why does the large model get dumber over time?

This phenomenon is usually caused by two dimensions: "context fatigue" within a single conversation and "model iteration and perception differences" during long-term use. You can resolve this by optimizing your prompts or checking relevant tutorials on platforms like Bilibili.

The following link will also update related tutorials from time to time, which can be watched and learned for free:
https://pan.quark.cn/s/d0f3910e647e

CheapAI Aggregation API novice tutorial & calling documentation | Support online calling/CherryStudio and other open source software calling/code calling

2. If you already have an API key, you can also use open-source software like CherryStudio on your computer or Chatbox on your phone for deployment and calling, to experience more powerful features. Configuration tutorials for common open-source software are as follows:#

3. Call via Code, please check the common document on the left#

4. AI Communication Group#

1. Welcome to log in to our newly created AI communication and learning website: CheapAI TOP https://cheapaiapi.store#