Jan
Open-source ChatGPT alternative that runs entirely offline on your computer with local models.
Open the official app on jan.ai
This tool is hosted by its maintainers. Click below to open jan.ai in a new tab — it's their official demo.
Browse text & tools tools →What's next with Jan?
Choose how you want to get started.
Use it free
Open the official tool or demo — no account needed.
Self-host it
Run the open-source version on your own infrastructure.
What is Jan?
Jan is an open-source AI tool designed as a privacy-focused alternative to commercial chatbots like ChatGPT. It enables users to run large language models (LLMs) locally on their computers, ensuring data remains offline and unshared with external servers. The tool caters to developers, designers, and knowledge workers seeking customizable AI assistance without compromising data security. By leveraging open-source models such as Llama, Mistral, and Qwen, Jan addresses the need for privacy-conscious AI solutions while offering flexibility in model selection. Its minimalist interface and community-driven development model make it accessible to both technical users and those with limited coding experience. Jan's offline operation and support for multiple AI models position it as a versatile tool for tasks ranging from coding to creative problem-solving.
How it works
Jan is an open-source project that provides local execution of AI models, eliminating reliance on cloud-based services. It functions as a customizable AI assistant, allowing users to integrate preferred models like Llama, Mistral, or Qwen directly on their devices. The tool prioritizes privacy by ensuring all computations occur offline, making it suitable for users concerned about data security. Its modular design enables developers to extend functionality through plugins and custom configurations. Jan supports a wide range of open-source and commercial models, including Llama, Mistral, Qwen, and Gemini, through plugin integration. Users can switch between models to tailor responses to specific tasks. Its offline operation ensures no data is transmitted to external servers, enhancing privacy.
How to use it
- 1Clone the Jan repository from GitHub: `git clone https://github.com/janhq/jan.git`.
- 2Install dependencies using the project's specified package manager.
- 3Configure model integration by selecting preferred LLMs via plugin setup.
- 4Launch the application and interact with the AI interface directly. Practical tips include using the terminal for advanced configurations and verifying model compatibility with system requirements.
What it can do
- offline AI assistant
Use cases
Assumptions and limitations
Assumptions
- source: https://github.com/janhq/jan
- license: AGPL-3.0 — free to use
- privacy: Self-hosted — you control your data
Limitations
- The tool's name may cause confusion with the unrelated personal name 'Jan'
- Limited documentation for advanced customization features
- Dependence on third-party models for core functionality
- Potential performance variations based on local hardware specifications
- Incomplete implementation of memory persistence features
Understanding the result
Open-source ChatGPT alternative that runs entirely offline on your computer with local models.
Tool details
- Clearly flagged when a network request is needed.
- No account, no sign-up, and no tracking of your content.
- Powered by (MIT).
- Built with
- (janhq/jan)
- License
- MIT
- Runs locally
- No — requires a network request
- Verification
- Not yet verified
- Input
- Query
- Output
- Text
Built with janhq/jan. OpenToolVault provides the discovery and browser interface while crediting the original project maintainers.
- Built with
- License
- MIT
Open-source project
OpenToolVault is an independent directory. We are not affiliated with or endorsed by this project.
References
- / — GitHub Repository
Upstream project · GitHub
- AGPL-3.0 License
Upstream project
Frequently asked
Why is the tool named 'Jan'?
The name 'Jan' is chosen for its simplicity and universal recognition as a common first name across multiple cultures. This choice aligns with the project's goal of creating an accessible, non-proprietary AI tool. However, the name may cause confusion with the unrelated personal name 'Jan', which is noted in the project's documentation.
How does Jan handle model compatibility?
Jan uses a plugin architecture to integrate with various models, but compatibility depends on the specific model's licensing and implementation. Users must verify that their selected model meets Jan's technical requirements and adheres to its open-source licensing terms. Some models may require additional dependencies or configuration adjustments.
How do I set up a custom model with Jan?
To add a custom model, first ensure it meets Jan's compatibility criteria. Then, modify the configuration file to include the model's parameters. Follow the project's documentation for steps to rebuild the application with the new model integration. Testing is recommended to confirm proper functionality.
How does Jan compare to ChatGPT?
Jan differs from ChatGPT by running entirely offline and using open-source models instead of proprietary ones. While ChatGPT requires cloud-based computation, Jan offers greater privacy through local execution. However, Jan's capabilities depend on the selected model's features, whereas ChatGPT provides a standardized experience across platforms.
What should I do if a model fails to load?
First, verify the model's compatibility with Jan's current version. Check for missing dependencies or incorrect file paths in the configuration. If the issue persists, consult the GitHub repository's issue tracker for potential solutions or report the problem to the community for assistance.