GoingMerry Documentation
Welcome to the GoingMerry developer documentation hub. GoingMerry is a local, performance-first model execution runtime compiled to run massive models directly on your hardware boundaries with bare-metal speed, air-gapped security, and zero cloud overhead.
What is GoingMerry?
GoingMerry is designed to provide local AI capabilities. It implements standard local execution APIs, allowing you to run, customize, build, and deploy local model workloads with high efficiency.
Documentation Categories
🚀 Core Guides & Setup
Get up and running with GoingMerry across macOS, Linux, Windows, or Docker. Learn about hardware allocation, Modelfiles, and the command-line interface.
🧠 Capabilities
Discover how to utilize advanced model capabilities including:
- Structured Outputs: Enforce JSON formatting.
- Tool Calling: Allow models to execute functions.
- Reasoning & Thinking: Control thinking steps in reasoning models.
- Vision: Process images.
🔌 API Reference
Developer documentation for the local REST API running on port 11434. Learn how to programmatically control inference, pull/push models, and configure embedding vectors.
🤝 Integrations
Guides on connecting GoingMerry to code editors, IDE plugins, desktop interfaces, and automation orchestrators (such as Claude Code, Cline, Zed, and n8n).