Local runtime - Menlo Research (janhq) - verified
Jan: install, API, platforms and features
Jan is a local LLM runtime from Menlo Research (janhq). As its docs describe it: Jan is an open-source desktop app for downloading and running local models offline on macOS, Windows and Linux, with an optional OpenAI-compatible local API server. The latest release is v0.8.6, published October 9, 2026. License: Apache-2.0 (open source).
- Latest release
- v0.8.6
- Released
- 2026-10-09
- License
- Open source
- OpenAI API
- Yes
Install
Copied verbatim from the page linked under each command. Labels are worded as the source words them.
macOS (Apple silicon)
Download Jan for Mac. Open the Jan installer (.dmg file), drag Jan to Applications, then launch Jan. Apple silicon is required for local model inference.
No copy-paste command is published for this platform. Source: docs.jan.ai/desktop/install/mac
Windows (x64)
Download Jan for Windows (x64). Run the downloaded .exe file, wait for installation to complete, then launch Jan. On a Windows on ARM PC, get the ARM64 installer from the download page.
No copy-paste command is published for this platform. Source: docs.jan.ai/desktop/install/windows
Run a model
The first-run example from the official docs. OpenAI-compatible endpoint: http://127.0.0.1:1337/v1
The official pages read for this directory publish no copyable run command for Jan. Models are downloaded and started from the desktop app.
Features
What the Jan docs say it supports. Not stated means the docs do not say either way.
| Feature | Value | Source |
|---|---|---|
| Interface | Desktop app, Server | See sources below |
| Platforms | macOS, Windows, Linux | docs.jan.ai/desktop/quickstart |
| Backends | CPU, CUDA, Vulkan, Metal | docs.jan.ai/desktop/local-engine/llama-cpp |
| Model formats | GGUF, MLX | docs.jan.ai/desktop/local-engine/mlx |
| OpenAI-compatible API | Yeshttp://127.0.0.1:1337/v1 | docs.jan.ai/desktop/api-server |
| Anthropic-compatible API | Not stated | Docs do not say |
| Tool calling | Yes | docs.jan.ai/desktop/mcp |
| Structured outputs | Yes | docs.jan.ai/desktop/local-engine/llama-cpp |
| Vision (image input) | Yes | docs.jan.ai/desktop/quickstart |
| Multi-GPU | Yes | docs.jan.ai/desktop/local-engine/llama-cpp |
| MCP client | Yes | docs.jan.ai/desktop/mcp |
| License | Apache-2.0Open source | github.com/janhq/jan/blob/dev/LICENSE |
| Latest release | v0.8.6October 9, 2026 | github.com/janhq/jan/releases/tag/v0.8.6 |
Models to run with Jan
Browse the open-weight models you can run locally for download sizes, licenses and memory fit; check which formats Jan loads in the table above.
Compare Jan
FAQ
Does Jan have an OpenAI-compatible API?
Yes. Endpoint, per its docs: http://127.0.0.1:1337/v1. Source: https://docs.jan.ai/desktop/api-server
Does Jan have an Anthropic-compatible API?
Its official docs, read on October 18, 2026, do not say.
Which platforms and GPUs does Jan support?
Its docs list macOS, Windows and Linux, with these backends: CPU, CUDA, Vulkan and Metal.
How do I install Jan?
On linux (Flatpak): "flatpak install flathub ai.jan.Jan", as shown at https://docs.jan.ai/desktop/install/linux. This page lists 1 install command: linux (Flatpak). For macOS (Apple silicon) and Windows (x64) the docs publish installer downloads instead of a command; both are linked in the install section.
Is Jan open source?
Yes. It is released under Apache-2.0; the license is at https://github.com/janhq/jan/blob/dev/LICENSE.
What is the latest version of Jan?
v0.8.6, published October 9, 2026 (https://github.com/janhq/jan/releases/tag/v0.8.6), as of October 18, 2026.
Which model formats does Jan run?
GGUF and MLX, per https://docs.jan.ai/desktop/local-engine/mlx
Sources
Last verified from the pages below.
- Jan GitHub README - github.com
- Jan releases (v0.8.6) - github.com
- Jan releases Atom feed - github.com
- Jan Desktop overview - docs.jan.ai
- Jan Quickstart - docs.jan.ai
- Local API Server - docs.jan.ai
- Local AI Engine (llama.cpp) - docs.jan.ai
- MLX engine - docs.jan.ai
- Model Context Protocol in Jan - docs.jan.ai
- Linux installation - docs.jan.ai
- Mac installation - docs.jan.ai
- Windows installation - docs.jan.ai
- Jan CLI and Desktop - docs.jan.ai
Newsletter
Local AI, every week
One email per week. New model releases, runtime updates and what they mean for what you build.
Keep exploring
Open-weight models you can run
Download sizes, licenses and the run command for each model.
All local LLM runtimes
Every runtime in one feature matrix, with sources.
Picking a runtime for coding agents
Ollama, LM Studio, vLLM and llama.cpp for agent workloads.
Setup guides
Point your coding tools at the model you picked.