GLM Vision

Integrates Z.AI's GLM-4.5V vision model to analyze images from local files or URLs with configurable parameters including temperature control, thinking mode for model reasoning exposure, and token limits up to 64K.

This MCP server by danilofalcao integrates GLM-4.5V vision model from Z.AI, providing image analysis capabilities through a single tool that supports both local files and URLs. Built with FastMCP and httpx for async HTTP operations, it features configurable parameters including temperature control, thinking mode for exposing model reasoning, and adjustable token limits up to 64K. The implementation includes comprehensive error handling, base64 encoding for local images, MIME type detection, and GitHub Actions workflows for CI/CD with PyPI publishing, making it valuable for developers who need AI-powered image analysis integrated into their workflows using Z.AI's GLM vision models rather than OpenAI's alternatives.