Jan
An open-source desktop chat application and local API for running models privately on a personal computer.
By Jan HQ · reviewed 2026-08-15
Deep profile checked 2026-08-16 · Jan desktop documentation and GitHub releases
What Jan does
Jan combines local model downloads, chat, assistants, an OpenAI-compatible server, and optional cloud providers in a desktop application. It is a strong alternative to LM Studio for users who prefer an open-source interface.
Pricing model
Free and open source
Position in the stack
Where Jan fits
Jan belongs primarily in the model runtimes layer. Jan combines local model downloads, chat, assistants, an OpenAI-compatible server, and optional cloud providers in a desktop application. It is a strong alternative to LM Studio for users who prefer an open-source interface. It should be evaluated as one part of a complete stack, because model files, inference providers, storage, identity, and external integrations remain separate operational choices.
Typical deployment flow
A practical sequence for evaluating Jan before making it part of a permanent stack.
- STEP 1
Confirm the deployment boundary
Choose among desktop-local based on users, data sensitivity, network access, and who will operate updates.
- STEP 2
Validate the complete stack
Check Desktop application, connected providers, supported formats, and exact hardware or accelerator compatibility before rollout.
- STEP 3
Run a representative workflow
Test open-source desktop local chat, record versions and settings, then review security, backups, observability, and failure recovery.
Best for
- Open-source desktop local chat
- Running GGUF models without command-line setup
- A local API and MCP-enabled assistant
Not the right layer for
- Shared multi-user serving
- Multi-node or production GPU scheduling
System fit
Requirements and hardware notes
Application requirements are separate from the memory needed by the selected model and context window.
Check your hardwareMinimum baseline
- macOS 12+, Windows 10+, or Ubuntu 20.04+
- 8GB RAM
- 10GB free storage before model downloads
Recommended baseline
- 16GB RAM
- 50GB free storage for several local models
- A supported Apple, NVIDIA, AMD, or Intel GPU
Hardware-specific notes
- Apple Silicon uses Metal automatically.
- Jan recommends choosing model files below roughly 80% of available RAM.
- Windows x64 acceleration and backend requirements vary by GPU vendor.
Installation and deployment paths
Choose one path that matches the number of users and the level of operations you can maintain.
Data boundary
What stays local and what may leave
- Local data
- Jan documents local storage for models, conversations, settings, and logs when using local models.
- Network dependencies
- Model downloads, web tools, MCP connectors, and configured cloud providers can transmit data outside the machine.
- Accounts and telemetry
- Evaluate each connector and cloud provider separately; a local desktop shell does not make every connected action local.
License checkpoints
- Jan is open source, but model and connector licenses are separate.
- Cloud providers configured in Jan retain their own terms and data policies.
Operational checkpoints
- Back up the Jan data directory before major upgrades.
- Port 1337 conflicts can prevent the local API from starting.
- Test MCP tools with least-privilege access before allowing file or account actions.
Capabilities
Capabilities refer to the tool's application layer. Hardware and model support can still depend on a connected inference engine.
Deployment and compatibility
- Deployment
- Desktop / local
- Platforms
- macOS · Windows · Linux
- Hardware backends
- CPU · Apple Silicon · NVIDIA GPU · AMD GPU · Intel GPU
- Install methods
- Desktop application
Models and integrations
- Model formats
- GGUF
- Common integrations
- Hugging FaceOpenAI-compatible clientsMCP servers
Strengths
- Open-source desktop interface
- Integrated local server
- Local and cloud provider support
Limitations
- Desktop-first rather than server-first
- Model support follows the bundled runtime
Planning checklist
Before you choose Jan
Answer these questions with the exact models, hardware, users, and data you expect to operate.
- Does Jan support the exact model, provider, data source, and operating system required by the workflow?
- Can the available hardware and memory handle the selected models, context, concurrency, and runtime overhead?
- Do the Apache-2.0 terms fit internal use, modification, redistribution, and any commercial service being planned?
- Who will own upgrades, credentials, backups, monitoring, and recovery when this tool becomes part of a real workflow?
Jan FAQ
What layer does Jan replace?
Jan primarily covers model runtimes. It does not automatically replace every model runtime, application, storage service, or infrastructure dependency connected to that layer.
Can Jan run entirely locally?
Yes, a local or self-hosted path is available. Privacy still depends on the model providers, connectors, telemetry, and external tools that you enable.
What should be tested before adopting Jan?
Use the exact models, documents, integrations, hardware, concurrency, and security boundary expected in production. Feature lists and public benchmarks cannot validate that complete combination.
Official sources
Use these links to confirm current compatibility and installation requirements.
Related tools
A polished desktop application for discovering, running, and serving local models.
A private desktop chat application and Python SDK for running GGUF language models on everyday computers.
A compact GGUF runtime and web interface focused on local text generation, roleplay, and story workflows.
Comparisons featuring Jan
Use a reviewed comparison when the choice is between two adjacent tools.
Execution evidence
Known working recipes using Jan
Recipes connect hardware, a model artifact, tools, settings, verification, and a reportable result.