介绍

官方网站：https://jan.ai/

Jan是一个开源的ChatGPT替代品，允许用户在其计算机上100%离线运行。这意味着用户无需网络连接即可使用AI功能。

软件截图

软件特点

• 离线运行：Jan支持在任何硬件上运行，从个人电脑到多GPU集群，无需外部服务器或网络连接。
• 多平台支持：提供Windows、MacOS和Linux的安装包，确保跨平台兼容性。
• 模型支持：用户可以根据需要下载并运行多种AI模型，包括Mistral、Llama等。
• 内置Hub：Jan内置了一个Hub模块，提供20多个大模型供用户选择，并支持手动导入其他模型。

更新日志

v0.6.10

Bring back QA changes 0.6.9 by @dinhlongviolin1 in #6320
docs: Major restructure and consolidation by @ramonpzg in #6323
hotfix: fixed home page hyperlink and extension gif by @ramonpzg in #6324
fix: Use 80% total memory for compatibility check (#6321) by @dinhlongviolin1 in #6333
hotfix: changed website theme by @ramonpzg in #6335
fix: mise build failing by @theproductiveprogrammer in #6344
chore: fix id code-block for avoid duplicate same state by @urmauur in #6346
chore: update docs font by @urmauur in #6359
fix: remove timezone specific from utils by @urmauur in #6363
fix: search hgf repo and downloaded filter by @urmauur in #6366
feat: allow see Apikey when server local status running by @urmauur in #6364
feat: download llamacpp backend fail back to cdn by @hiento09 in #6361
feat: Jan Web (reusing Jan Desktop UI) by @dinhlongviolin1 in #6298
feat: Web use jan model by @dinhlongviolin1 in #6374
ci: add paths for jan server web by @Minh141120 in #6377
chore: update Dockerfile to use custom nginx.conf by @Minh141120 in #6378
docs: add first‑class API Reference to Jan docs (Local + Server) by @ramonpzg in #6372
ci: remove unnecessary folder paths and on Dockerfile by @Minh141120 in #6392
enhancement: responsive dialog modals by @urmauur in #6381
feat: allow user import model include mmproj file by @urmauur in #6388
chore: add cicd for jan web prod by @hiento09 in #6396
feat: improve testing by @dinhlongviolin1 in #6395
fix: jan server api base url for prod by @hiento09 in #6403
removed cloud api spec update by @ramonpzg in #6406
ci: add claude issue dedup by @Minh141120 in #6408
enhancement: rollback edit capabilities for local model by @urmauur in #6409
feat: Smart model management by @qnixsynapse in #6390
feat: enhance llamacpp backend management and installation by @qnixsynapse in #6360
ci: add nightly build for external contributors by @Minh141120 in #6413
fix: include lm_head and embedding layers in totalLayers count by @qnixsynapse in #6415
fix: clean up unused packages by @louis-menlo in #6414
fix: render new line for user message by @urmauur in #6412
bug: Deleted model file from imported models blocking model loading (#6317) by @dinhlongviolin1 in #6417
enhancement: add label experimental for optimize setting by @urmauur in #6416
fix: error connecting to ollama (#6349) by @dinhlongviolin1 in #6418
exclude jan extension web from desktop build by @dinhlongviolin1 in #6419
fix: avoid error validate nested DOM by @urmauur in #6426
fix: correct context shift flag handling in LlamaCPP extension (#6404) by @qnixsynapse in #6431
enhancement: copy MCP permission by @urmauur in #6456
fix: KVCache size calculation and refactor by @qnixsynapse in #6438
chore: makefile rust targets macos by @Minh141120 in #6457
fix: detect allocation failures as out-of-memory errors by @qnixsynapse in #6459
fix: set default memory mode and clean up unused import by @qnixsynapse in #6463
fix: pathname file install BE by @urmauur in #6464
fix: attachment edit message by @urmauur in #6465
fix: immediate update value model selection by @urmauur in #6474
fix: validate mmproj from general basename by @urmauur in #6477
fix: new extension settings aren't populated properly by @louis-menlo in #6476
chore: embed webview2 bootstrapper in tauri windows by @Minh141120 in #6482
fix: mcp json validation structure and serverName by @urmauur in #6478
chore: prevent click outside for edit dialog by @urmauur in #6493
feat: add qa checklist v0.6.10 by @Minh141120 in #6495
fix: correct memory suitability checks in llamacpp extension by @qnixsynapse in #6504

下载地址

本地下载        蓝奏云        123云盘