0.5.13

🐛 Fixes

• fix: max_tokens revert back to 8192 automatically when creating a new thread @louis-jan (#4397)
• fix: revert back product analytic with default opt-out @urmauur (#4394)
• fix: send event opt out @urmauur (#4390)
• fix: enable default opt in analytic @urmauur (#4387)
• fix: should not disable Vulkan support option @louis-jan (#4382)
• fix: bump llama.cpp engine hotfix version to address model load issue on some MacOS Intel machines @louis-jan (#4374)

🧰 Maintenance

• chore: bump latest cortex release 1.0.8 @louis-jan (#4405)
• chore: add cpu_threads settings in cortex extension @louis-jan (#4386)

Contributor

@louis-jan and @urmauur

0.5.13-rc5-beta

What's Changed

• fix: max_tokens revert back to 8192 automatically when creating a new thread by @louis-jan in #4397

Full Changelog: v0.5.13-rc4-beta...v0.5.13-rc5-beta

0.5.13-rc4-beta

What's Changed

• fix: send event opt out by @urmauur in #4390
• fix: revert back product analytic with default opt-out by @urmauur in #4394

Full Changelog: v0.5.13-rc2-beta...v0.5.13-rc4-beta

0.5.13-rc3-beta

What's Changed

• fix: send event opt out by @urmauur in #4390

Full Changelog: v0.5.13-rc2-beta...v0.5.13-rc3-beta

0.5.13-rc2-beta

What's Changed

• fix: should not disable Vulkan support option by @louis-jan in #4382
• chore: add cpu_threads settings in cortex extension by @louis-jan in #4386
• fix: enable default opt in analytic by @urmauur in #4387

Full Changelog: v0.5.13-rc1-beta...v0.5.13-rc2-beta

0.5.13-rc1-beta

What's Changed

• fix: bump llama.cpp engine hotfix version to address model load issue on some MacOS Intel machines by @louis-jan in #4374

Full Changelog: v0.5.12...v0.5.13-rc1-beta

0.5.12

🚀 Features

• enhancement: set recommend model to undefined if no model available @urmauur (#4296)
• enhancement: update thumbnail chat width option setting @urmauur (#4284)
• feat: handle case thread when original model deleted @urmauur (#4280)
• feat: add chat width setting option for thread @urmauur (#4278)
• feat: reroute threads and messages requests to cortex.cpp backend @louis-jan (#4249)
• feat: product analytic UI @urmauur (#4262)
• enhancement: move token speed to bottom @urmauur (#4276)
• feat: add cicd for janhq/joi @hiento09 (#4270)
• enhancement: better error handing for remote models when there's no internet connection @urmauur (#4252)
• feat: add auto build when PR is approved @hiento09 (#4241)
• enhancement: default open Jan Input Box Settings and Right panel @urmauur (#4234)
• feat: update app download universal for macos @urmauur (#4230)
• enhancement: remove tooltip local api server when server start @urmauur (#4275)

🐛 Fixes

• fix: button state reset on switching models @louis-jan (#4356)
• fix: regenerate scroll bottom @urmauur (#4348)
• fix: send message button state reset on stop @louis-jan (#4341)
• fix: render performance while generating messages @louis-jan (#4328)
• fix: disabled native resize textarea when textarea autogrowing @urmauur (#4326)
• fix: scroll bottom when generation text @urmauur (#4323)
• fix: app picks model automatically edge cases @louis-jan (#4322)
• fix: generating message is gone after switching to another thread @louis-jan (#4321)
• fix: update thread's title edge cases @louis-jan (#4320)
• fix: threads sorting order after updated @louis-jan (#4319)
• fix: change theme settings do not work after relocated data folder @louis-jan (#4317)
• fix: cleaning a thread should just clear out messages @louis-jan (#4316)
• fix: dynamic input box width and remove tooltip image option @urmauur (#4315)
• fix: overlap bottom system monitor @urmauur (#4310)
• fix: scroll bottom issue @urmauur (#4308)
• fix: fetch threads and messages without limiting @louis-jan (#4304)
• fix: thread error handling @louis-jan (#4307)
• fix: escape heading on user message item @urmauur (#4301)
• fix: conditional render model setting based on selected model @urmauur (#4300)
• fix: attach file information into message metadata for a quick retrieval @louis-jan (#4298)
• fix: new thread isn't automatically created on factory reset @louis-jan (#4295)
• fix: scroll to bottom issue @urmauur (#4289)
• fix: context length is reset while creating a new thread @louis-jan (#4294)
• fix: empty tooltip on attachment icon, also fix z-index @urmauur (#4290)
• fix: attachment preview issue @louis-jan (#4282)
• fix: performance issue with atom storage persistence @louis-jan (#4281)
• fix: starter screen layout @urmauur (#4279)
• fix: disabled start model from mymodel list when server is running @urmauur (#4277)
• fix: default max_tokens set on remote models @louis-jan (#4266)
• fix: markdown inline inputbox @urmauur (#4269)
• fix: auto scrolling to bottom @urmauur (#4256)
• fix: crash markdown render code block without triple backtick @urmauur (#4248)
• fix: performance issue when pasting long content into the chat input box @urmauur (#4240)
• fix: hide Vulkan option for users who don't have GPU @louis-jan (#4226)
• fix: list style off screen @urmauur (#4231)
• fix: fix issue with duplicate events on Enter key press during CJK character composition in chat input @cuhong (#4268)

🧰 Maintenance

• chore: bump cortex latest version 1.0.7 @louis-jan (#4357)
• chore: bump cortex 1.0.7 RC @louis-jan (#4355)
• chore: don't stop inference and hide the stop button when creating a new thread @louis-jan (#4354)
• chore: Refactor thread state management @louis-jan (#4350)
• chore: bump latest cortex stable release version @louis-jan (#4336)
• chore: bump cortex 1.0.6-rc @louis-jan (#4330)
• chore: add OpenAI gpt-4o-mini and o1 models @louis-jan (#4311)
• chore: deprecate Jan core's REST module - all API requests go to cortex.cpp @louis-jan (#4297)
• chore: bump cortex 1.0.5 rc2 @louis-jan (#4293)
• chore: add ci janhq/core publish npm @hiento09 (#4259)
• chore: deprecate groq llama 3.1 70B Versatile @louis-jan (#4265)
• chore: fix openai vision models @louis-jan (#4246)
• chore: roadmap template fix @gabrielle-ong (#4227)
• chore: fix beta ci @hiento09 (#4233)
• chore: update download url for macos universal @hiento09 (#4232)

Contributor

@cuhong, @gabrielle-ong, @hiento09, @louis-jan and @urmauur

0.5.12-rc7-beta

What's Changed

• fix: button state reset on switching models by @louis-jan in #4356

Full Changelog: v0.5.12-rc6-beta...v0.5.12-rc7-beta

0.5.12-rc6-beta

What's Changed

• chore: bump cortex 1.0.7 RC by @louis-jan in #4355

Full Changelog: v0.5.12-rc5-beta...v0.5.12-rc6-beta

0.5.12-rc5-beta

What's Changed

• chore: don't stop inference and hide the stop button when creating a new thread by @louis-jan in #4354

Full Changelog: v0.5.12-rc4-beta...v0.5.12-rc5-beta