modelgrid/changelog.md

Changelog

2026-01-30 - 1.0.0 - Initial Release

ModelGrid v1.0.0 - GPU Infrastructure Management Daemon

ModelGrid is a root-level daemon that manages GPU infrastructure, Docker containers, and AI model serving with an OpenAI-compatible API interface.

Features

GPU Management:

• Multi-vendor GPU detection (NVIDIA/CUDA, AMD/ROCm, Intel Arc/oneAPI)
• Automatic driver detection and installation guidance
• Real-time GPU status monitoring
• GPU assignment to containers

Container Management:

• Support for Ollama, vLLM, and TGI (Text Generation Inference) containers
• Automatic container lifecycle management
• GPU passthrough configuration
• Health monitoring

OpenAI-Compatible API:

• /v1/chat/completions - Chat completions with streaming support
• /v1/models - List available models
• /v1/embeddings - Text embeddings
• Bearer token authentication

Model Management:

• Greenlit model system for controlled auto-pulling
• Automatic model loading on request
• VRAM requirement validation
• Model registry from remote configuration

System Integration:

• Systemd service management
• Configuration at /etc/modelgrid/config.json
• Comprehensive CLI for all operations

Installation

# Via npm (recommended)
npm install -g @modelgrid.com/modelgrid

# Via installer script
curl -sSL https://code.foss.global/modelgrid.com/modelgrid/raw/branch/main/install.sh | sudo bash

Supported Platforms

• Linux x64 (x86_64)
• Linux ARM64 (aarch64)
• macOS Intel (x86_64)
• macOS Apple Silicon (ARM64)
• Windows x64

CLI Commands

modelgrid service enable|disable|start|stop|status|logs
modelgrid gpu list|status|drivers
modelgrid container add|remove|list|start|stop
modelgrid model list|pull|remove
modelgrid config show|init