The image/video generation API provided by ZhipuAI, support cogview and cogvideox models.
CogView
Overview
ZhipuAI is an AI model for text-to-image and text-to-video generation. It can create both images and videos based on textual descriptions provided by users. The model offers a variety of artistic styles and content generation options across both image and video formats.
Configuration
1. Ensure that you are allowed to use the BigModel service
If you don't have an account, go to the BigModel site and follow their instruction to create an account.
You can get the API key from the ZHIPU AI OPEN PLATFORM.
2. Get CogView tools from Plugin Marketplace
The ZhipuAI tools could be found at the Plugin Marketplace, please install it.
3. Setup the CogView API in the Dify Tool page
On the Dify navigation page, click Tools > zhipuai > To Authorize and fill the API Key.
4. Using the tool
You can use the ZhipuAI tool in the following application types.
Chatflow / Workflow applications
Both Chatflow and Workflow applications support adding a ZhipuAI tool node.
Agent applications
Add the ZhipuAI tool in the Agent application, then enter online search instructions to call this tool.