Our take
Activeregular commits and releases- Activity
- 77
- Community
- 91
- Issues
- 89
- Pull requests
- 81
+767 stars in the last 30 days
- Merges more pull requests than 90% of the projects we track
- Closes issues faster than 81% of the projects we track
- More contributors than 78% of the projects we track
Jan is an actively maintained, cross-platform desktop option for people who want local model access alongside cloud connections, backed by a broad contributor base and frequent releases.
Good fit if
- You want a desktop interface for trying local models and can meet the model’s hardware needs.
- You need a local OpenAI-compatible endpoint or want to combine local and cloud model access.
- You value a project with recent commits and releases and contributions from many people.
Look elsewhere if
- You need a hosted, centrally managed service rather than software running on users’ computers.
- Your workflow requires models or integrations beyond the local-model, listed cloud-provider, and MCP capabilities described here.
- You cannot meet the Node.js and Rust prerequisites needed to build from source, or do not want to install a desktop application.
All health signals
| Last commit | 2026-10-08 (today) |
|---|---|
| Commits, last 90 days | 100+ |
| Releases, last 12 months | 14 (latest v0.8.5, 2026-10-08) |
| Contributors | 100+ (top contributor: 22% of commits) |
| Issues closed, last 90 days | 22+ (typically closed in 0 days) |
| Pull requests merged, last 90 days | 85 (typically merged in 0 days) |
| Project age | 3 years |
Checked on 2026-10-09 with the GitHub API.
Project history
Daily GitHub snapshots since Oct 7, 2026. History covers stars and forks; health signals are not tracked historically.
Stars
45,716Forks
3,171Change since Nov 13, 2026: +21 stars·0 forks
Overview
Jan is a desktop app for running and chatting with language models on your computer. It addresses the need for a familiar chat interface that can work with local models, so using AI does not always require sending prompts to a cloud service.
You can also connect cloud providers when local models are not the right fit. That combination makes Jan useful for individuals who want to experiment with local AI, and for developers who need a local model endpoint alongside a desktop interface.
Key Features
- Download and run local models, including models from Hugging Face.
- Connect to cloud models from providers such as OpenAI, Anthropic, Mistral, and Groq.
- Create custom assistants for different tasks.
- Expose a local, OpenAI-compatible API at
localhost:1337. - Integrate Model Context Protocol (MCP) for agentic capabilities.
- Install on Windows, macOS, or Linux.
Use Cases
- Privacy-conscious individuals can chat with local models when they prefer not to send prompts to a cloud provider.
- Developers can test applications against Jan’s local OpenAI-compatible API.
- People comparing models can use local models and connect cloud providers in the same desktop app.
- Teams or individuals can create task-specific assistants and use MCP integrations for agentic workflows.
What you need
Detected in the repository
- Node.js (from package.json)
- A test suite and automated checks on GitHub Actions
Getting Started
Download an installer for your platform from jan.ai or GitHub Releases. See the README for build instructions and platform details.
License in Plain Words
No standard license was detected. Without a license, the default copyright rules apply: you may read the code, but not reuse it. Check the repository or ask the authors.
Alternatives
- cherry-studio: Cherry Studio emphasizes model comparison, ready-made assistants, and document workflows alongside local and cloud chat.
- deepchat: DeepChat adds agents, tools, skills, and resumable sessions to local and cloud model chat.
| Project | Language | License | Stars | Status |
|---|---|---|---|---|
| jan | Rust | – | 44.9k | Active |
| cherry-studio | TypeScript | AGPL-3.0 | 52.7k | Active |
| deepchat | TypeScript | Apache-2.0 | 6.4k | Active |
Considerations
Local inference depends on your computer’s memory and hardware, and model sizes can make the experience resource-intensive. The README gives different memory guidance for macOS model sizes and describes GPU support for Windows and Linux; it does not promise that every model will run well on every system. Building from source requires Node.js, Yarn, Make, and Rust, with additional platform-specific setup. The project is active, with regular commits and releases and contributions from at least a hundred contributors, but that activity does not by itself establish production suitability for a particular deployment. The README says the license is Apache 2.0, while the supplied GitHub license metadata is listed as NOASSERTION, so verify the applicable license before adoption.
Found this useful?
Share it with someone who would like jan.