[deepin exploration] UOS AI Integrates Online Models EN
Tofloor
poster avatar
deepin小助手
Super Moderator
OM
2024-07-05 16:36
Author

The UOS AI version 1.3.0 has been released in the V23 App Store, and the new version now supports adding custom models!

Access Conditions

The interface specification for custom models in UOS AI follows OpenAI's /v1/chat/completions. Therefore, only model services that provide an OpenAI-compatible interface can be added to UOS AI.

Online Models

Next, we will introduce how to add an online model.

OpenAI

● Obtain API

Open OpenAI's API documentation: https://platform.openai.com/docs/api-reference/chat/create

Request URL: https://api.openai.com/v1/chat/completions

Model Name: gpt-4o

Choose a model based on your needs, such as gpt-4o, gpt-4, or gpt-3.5-turbo. Model list: https://platform.openai.com/docs/models

● Obtain API Key

Log in to the OpenAI console, go to "Your profile," and select "User API keys": https://platform.openai.com/settings/profile?tab=api-keys

Click the "Create new secret key" button to generate an API Key.

Copy the generated key from this interface.

● Add Model to UOS AI

Enter the UOS AI settings interface to add a model. In the model addition interface, switch the LLM option to “Custom.” Then fill in the following information:

○ API Key

Paste the key copied in the previous step into the API Key field.

○ Account Name

Fill in any name.

○ Model Name

Enter the model name from the first step's API documentation: gpt-4o

○ Request URL

UOS AI will automatically add /chat/completions to the request URL. Therefore, the address entered here should be the OpenAI address without the /chat/completions part.

The actual address to enter is: https://api.openai.com/v1

Click confirm, and after verification, you can start using it in the chat window.

Reply Favorite View the author
All Replies

No replies yet