Our take
Activeregular commits and releases- Activity
- 88
- Community
- 31
- Issues
- 43
- Pull requests
- 79
+10 stars in the last 15 days
- Releases more often than 92% of the projects we track
- Merges more pull requests than 87% of the projects we track
OGAD is a feature-rich local AI desktop and gateway with strong recent development activity, but its open core is still relatively young and development is highly concentrated in one contributor.
Good fit if
- You want local inference across several modalities and value keeping conversations and files on-device.
- You need an OpenAI-compatible local endpoint, including a headless mode for your own applications.
- You are comfortable with AGPL-3.0 obligations and evaluating a fast-moving project.
Look elsewhere if
- You need a cloud-hosted service, managed accounts, or inference without running models on your own hardware.
- You need the Pro memory and activity features to be part of the open repository; they are distributed separately and license-gated.
- Your adoption requires development to be spread across a broad contributor base; most code contributions come from one person.
All health signals
| Last commit | 2026-10-12 (today) |
|---|---|
| Commits, last 90 days | 100+ |
| Releases, last 12 months | 86 (latest v0.0.55-beta.114, 2026-10-05, pre-release) |
| Contributors | 7 (top contributor: 85% of commits) |
| Issues closed, last 90 days | 3+ (typically closed in 19 days) |
| Pull requests merged, last 90 days | 79 (typically merged in 0 days) |
| Project age | 9 months |
Checked on 2026-10-13 with the GitHub API.
Overview
OGAD is a desktop application for running open AI models locally, with a studio for chat and media tasks and a local OpenAI-compatible API for other tools. It addresses the need to use models across text, vision, image, and audio workflows without sending conversations, files, or inference requests to a service provider.
The open core includes chat, document search, local model management, and MCP connectors. An optional Pro edition adds opt-in screen-based memory and other always-on features; those features are in a separate private package, not this repository's open core.
Key Features
- Run local text and vision models, including GGUF models, through bundled inference runtimes.
- Generate images and use on-device speech-to-text and text-to-speech.
- Organize conversations into projects and retrieve information from uploaded documents with cited sources.
- Connect MCP servers and invoke local tools from chat.
- Serve chat, image, audio, and embedding endpoints through a local OpenAI-compatible API.
- Run the gateway headlessly and manage models over its HTTP endpoints.
- Render generated HTML, React, SVG, Mermaid, and Markdown artifacts in a sandboxed view.
Use Cases
- A privacy-conscious individual can chat with local models and search personal documents without cloud inference.
- A developer can point an OpenAI-compatible client at the local gateway to experiment with local models in an existing workflow.
- A small team or homelab operator can run the gateway without the desktop UI and serve local models to other applications.
- Someone who needs multimodal work on one machine can combine local chat, image generation, transcription, and speech output in one studio.
What you need
Detected in the repository
- Node.js >=22 <23 (from package.json)
- A test suite and automated checks on GitHub Actions
License in plain words
AGPL-3.0network copyleft
- Commercial use: yes
- Modify and redistribute: yes
- You must keep: the license notice
- Share your changes: yes, the full source, also when people only use it over a network (e.g. as a hosted service)
- Includes an explicit patent grant from the contributors:
A summary, not legal advice: the LICENSE file is what applies.
Getting Started
Download a platform build from the latest releases. To build from source, install Node.js 22 and Git LFS, then:
git clone https://github.com/off-grid-ai/OGAD.git
cd OGAD
git lfs install && git lfs pull
npm install
npm run dev
See the README for platform-specific build instructions and setup details.
Alternatives
| Project | Language | License | Stars | Status |
|---|---|---|---|---|
| OGAD | TypeScript | AGPL-3.0 | 125 | Active |
| Jan | Rust | – | 45k | Active |
| textgen | Python | AGPL-3.0 | 47.7k | Active |
Considerations
This is a young project, though its measured health is active: it has regular commits and releases, with many recent merged pull requests. The contributor base is small and highly concentrated in one person, so maintenance continuity is a consideration. The available health data also shows relatively few issues closed in the recent period; that is worth weighing alongside the reported median time to close.
Inference runs on your machine, so model size and performance depend on local hardware. The repository describes CPU and Vulkan availability in addition to CUDA for Linux, rather than requiring a GPU in every configuration. Building requires Node.js 22, and native runtimes and platform-specific build steps add setup complexity. The core is AGPL-3.0-only, so review its terms before incorporating or distributing it; Pro features are separate and license-gated.
Found this useful?
Share it with someone who would like OGAD.