{"full_name":"janhq/jan","owner":"janhq","name":"jan","url":"https://github.com/janhq/jan","description":"Jan is an open source alternative to ChatGPT that runs 100% offline on your computer.","homepage":"https://jan.ai/","topics":[],"language":"Rust","license":"NOASSERTION","avatar":"https://avatars.githubusercontent.com/u/102363196?v=4","created_at":"2023-08-17T02:17:10Z","pushed_at":"2026-11-10T09:10:09Z","metrics":{"stars":45648,"forks":3168,"watchers":228,"issues":4102,"prs":4922,"releases":108,"issues_open":null,"issues_closed":null,"prs_open":null,"prs_closed":null,"prs_merged":null,"discussions":283},"growth":{"day":{"stars":27,"forks":2,"base_day":"2026-11-09"}},"history":{"dates":["2026-11-09","2026-11-10"],"stars":[45621,45648],"forks":[3166,3168],"watchers":[228,228],"issues_open":[null,null],"prs_open":[null,null]},"boards":{},"same_owner":[],"issues":[{"number":7728,"title":"bug: Jan does not download models in version v0.7.8","comments":32,"created_at":"2026-03-12T19:38:15Z"},{"number":7558,"title":"bug:  MLX in latest version 0.7.7 Broken and can not get it to work.","comments":20,"created_at":"2026-02-26T01:12:48Z"},{"number":8373,"title":"Build error: App built from tarball has GUI broken","comments":18,"created_at":"2026-07-01T19:29:01Z"},{"number":6017,"title":"discussion: No default backend included","comments":16,"created_at":"2025-07-31T18:46:37Z"},{"number":6166,"title":"idea: Voice input using local / remote whisper","comments":14,"created_at":"2025-08-13T13:26:42Z"},{"number":8266,"title":"bug: AMD ROCm GPU not recognized in flatpak linux","comments":12,"created_at":"2026-06-02T15:55:01Z"},{"number":7461,"title":"bug: Unable to start default Jan model in macOS 15.7.3","comments":12,"created_at":"2026-02-04T21:34:02Z"},{"number":8530,"title":"Bug: Thinking mode intermittent — reasoning block appears inconsistently even with mode set to ON or AUTO, regenerating fixes it","comments":11,"created_at":"2026-07-25T01:37:19Z"}],"discussions":[{"number":9620,"title":"How do I install REA or Universal modders?","comments":0,"category":"General","created_at":"2026-11-07T11:53:30Z"},{"number":9528,"title":"Multimodal RAG","comments":0,"category":"Feature Ideas","created_at":"2026-11-02T08:26:09Z"},{"number":9512,"title":"Podcast research in Jan: choose the file scope before adding your notes","comments":0,"category":"General","created_at":"2026-10-30T09:01:34Z"},{"number":9433,"title":"Manually unload a model","comments":2,"category":"Get Help","created_at":"2026-10-24T03:17:10Z"},{"number":9417,"title":"It would be great if Jan.ai didn't require an internet connection to use local tools/MCP","comments":1,"category":"Get Help","created_at":"2026-10-21T19:08:11Z"},{"number":9370,"title":"Is there a way to disable JAN's ability to run terminal commands?","comments":1,"category":"Get Help","created_at":"2026-10-20T18:53:25Z"},{"number":9309,"title":"Concise user guide in French","comments":0,"category":"Get Help","created_at":"2026-10-17T21:00:26Z"},{"number":9238,"title":"Option to set or pick a language for spellchecking.","comments":0,"category":"Feature Ideas","created_at":"2026-10-12T14:28:54Z"}],"latest_release":{"name":"0.8.8","tag":"v0.8.8","published_at":"2026-10-28T03:31:54Z","body":"## Jan v0.8.8: security fixes, llama.cpp 0.7.0, and Trusted Hosts for the local API\n\nA security update to v0.8.7. We recommend updating. Your data and settings carry over; a few behaviours change, listed under **What you might notice**.\n\n### Security\n\n- **Links in chat open in your browser.** Links in model replies, tool results, model cards and release notes now always open in your system browser, never inside Jan's window. Links that aren't web or email links are shown as plain text.\n- **The local API server checks where browser requests come from.** A web page can only call it if its origin is trusted: the same machine, Jan itself, or a host you list in **Trusted Hosts**. Requests without a browser origin (SDKs, scripts, curl, Claude Code) work as before.\n- **Server-side tool execution stays opt-in for every endpoint.** Requests to the local API run MCP tools only when **Settings > Local API Server** allows server-side tool execution.\n- **Paths are checked before Jan reads or writes them.** Thread and assistant ids, and the file operations extensions use, stay inside Jan's data folder.\n- **The Claude Code integration writes your shell profile safely.** Values are quoted, unusual variable names are refused, and Jan's lines are kept in one marked block that Reset removes completely without touching your own lines.\n- **The agent's file tools no longer follow a link out of their workspace**, and folders such as your home, credentials and startup folders can't be attached as writable.\n- **Windows: Jan runs system tools from the Windows system folder**, and sign-in links are opened without going through `cmd`.\n- **MCP sign-in only uses web addresses** (`http`/`https`) for its endpoints.\n\n### What you might notice\n\n- **Trusted Hosts for browser-based clients.** A web app on another machine (for example Open WebUI's direct connection) or a browser extension calling Jan's API needs its host added to **Settings > Local API Server > Trusted Hosts**. Entries can include a scheme, so `app://obsidian.md` works for Obsidian plugins (thanks to [@SnowingFox](https://github.com/SnowingFox), [#8608](https://github.com/janhq/jan/issues/8608)). Pages without a normal origin (local files, sandboxed frames) can be allowed by listing `file://` or `null`.\n- **The CORS switch now takes effect**, and it starts **on** after updating. In earlier versions it was shown but had no effect.\n- **Custom environment variables for Claude Code:** keys that change how programs are found (`PATH`, `PATHEXT`, `SystemRoot`, `ComSpec`, `LD_*`, `DYLD_*`, `JAN_*` and the GPU runtime paths Jan uses) are refused with a message. Other variables, including `NODE_OPTIONS`, save as before.\n- **Moving the data folder** to a system folder, your home folder itself or a drive root (for example `D:\\`) is refused; choose a folder such as `D:\\Jan`.\n- **MCP servers whose sign-in endpoints aren't web addresses** now show a sign-in error.\n\n### New\n\n**llama.cpp engine 0.7.0.**\n\nThe bundled engine moves from 0.5.0 to 0.7.0 (b11669). New per-model settings: an **MoE expert GPU cache** for mixture-of-experts models, **Probabilistic draft sampling** and **Draft batch size** for speculative decoding. They default to the engine's own defaults, so nothing changes unless you set them. Saved prompt caches from the previous engine are discarded once and rebuilt. ([#9399](https://github.com/janhq/jan/pull/9399))\n\n### Fixes\n\n**Local models failed to load on Windows when the user name had non-English characters** (for example Cyrillic). ([#9418](https://github.com/janhq/jan/pull/9418))\n\n**With an HTTP proxy configured, documents with \"Use embeddings\" failed**, because Jan's requests to its own engine went through the proxy. They now go direct; remote providers still use the proxy. ([#9353](https://github.com/janhq/jan/pull/9353))\n\n**An attachment was sent again on every turn**, adding a copy per message and defeating the prompt cache. It's now included once per request. Thanks to [@yaodong-shen](htt"},"readme":true,"readme_zh":true,"news":[],"stale":false,"snapshot_at":"2026-11-10T09:35:25Z","history_start":"2026-09-06"}