This repository provides a comprehensive collection of examples and guides for utilizing GLM APIs, targeting developers looking to integrate advanced AI capabilities into their applications. It offers practical code snippets, tutorials, and resources to facilitate the adoption of GLM's multimodal and agent-based functionalities.

How It Works

The cookbook primarily uses Python and Jupyter Notebooks to demonstrate GLM API usage, covering basic API calls, vision and multimodal models, fine-tuning, agent systems, and data analysis. The examples are structured into categorized folders for easy navigation, enabling users to quickly find relevant code for specific tasks like video understanding, multi-tool calling, and GraphRAG.

Quick Start & Requirements

• Install dependencies: pip install -r requirements.txt
• Recommended Python versions: 3.9 - 3.12.
• Requires a GLM API account and API key.
• Official SDKs available for Python, Java, C#, and Node.js.

Highlighted Details

• Features tutorials for video understanding, multi-tool calling, and GraphRAG.
• Includes examples for GLM-4V object recognition and agent system development.
• Demonstrates practical applications such as CSV data analysis and OCR integration.
• Offers guides for Zhipu AI agent API calls and third-party API integration.

Maintenance & Community

The repository is actively maintained by MetaGLM, with recent updates including video generation and understanding tutorials. SDKs for multiple languages have been released. Contributions are welcomed via Pull Requests and Issues.

Licensing & Compatibility

The repository's licensing is not explicitly stated in the provided text, but the availability of open-source SDKs suggests a permissive approach. Compatibility for commercial use would require verification of the specific license.

Limitations & Caveats

While the cookbook covers a wide range of GLM API features, some advanced functionalities or specific model integrations might require further exploration or custom implementation. The primary focus is on Python, with other languages requiring manual adaptation.