Jan, an open-source offline alternative to ChatGPT, has been experiencing significant user frustration due to model compatibility and performance issues following updates. The project, developed in TypeScript, aims to provide a self-hosted AI experience across various hardware configurations without relying on cloud services.
The most notable activity within the past 30 days includes a surge in user-reported issues, particularly concerning model loading failures and UI inconsistencies. The transition to the Cortex backend appears to have introduced instability, leading to numerous bug reports and feature requests aimed at improving user experience. Despite these challenges, the development team remains active, with ongoing efforts to enhance functionality through UI improvements and bug fixes.
Recent issues predominantly revolve around bugs related to model handling and user interface functionality. For instance, #3410 addresses missing placeholders for empty search results, while #3409 highlights import errors on Windows systems. These issues indicate a trend of technical difficulties affecting user satisfaction.
The development team has been actively working on resolving these issues and enhancing the application. Key contributors include:
Faisal Amir (urmauur)
Louis (louis-jan)
Van Pham (Van-QA)
NamH (namchuai)
hiento09
Gabrielle Ong (gabrielle-ong)
Mark Nguyen (marknguyen1302)
Developer | Avatar | Branches | PRs | Commits | Files | Changes |
---|---|---|---|---|---|---|
Louis | 6 | 24/22/2 | 34 | 355 | 82808 | |
NamH | 4 | 27/26/2 | 23 | 398 | 54938 | |
Faisal Amir | 7 | 46/39/5 | 46 | 96 | 3358 | |
hiento09 | 2 | 3/2/1 | 5 | 15 | 2473 | |
Mark Nguyen (marknguyen1302) | 1 | 1/0/0 | 2 | 12 | 1214 | |
Hien To | 2 | 0/0/0 | 2 | 9 | 341 | |
Van Pham | 2 | 4/4/0 | 4 | 3 | 78 | |
Nicole Zhu | 2 | 1/1/0 | 3 | 6 | 74 | |
Van QA | 1 | 0/0/0 | 2 | 3 | 6 | |
Gabrielle Ong | 2 | 1/1/0 | 2 | 1 | 4 | |
Luke Niesink | 1 | 1/1/0 | 1 | 1 | 2 | |
netwjx (netwjx) | 0 | 0/0/2 | 0 | 0 | 0 | |
None (Realmbird) | 0 | 0/0/1 | 0 | 0 | 0 | |
Quan Pham (Doko-Demo-Doa) | 0 | 1/0/1 | 0 | 0 | 0 | |
Marco Di Francesco (MarcoDiFrancesco) | 0 | 1/0/1 | 0 | 0 | 0 |
PRs: created by that dev and opened/merged/closed-unmerged during the period
Timespan | Opened | Closed | Comments | Labeled | Milestones |
---|---|---|---|---|---|
7 Days | 35 | 18 | 1 | 1 | 5 |
30 Days | 120 | 87 | 119 | 1 | 5 |
90 Days | 243 | 149 | 318 | 4 | 7 |
All Time | 1644 | 1451 | - | - | - |
Like all software activity quantification, these numbers are imperfect but sometimes useful. Comments, Labels, and Milestones refer to those issues opened in the timespan in question.
The recent activity on the Jan repository indicates a significant influx of issues, with 193 open issues currently documented. The majority of these issues are categorized as bugs, particularly related to model loading and user interface functionality. Notably, many users report problems with model compatibility and performance, especially after updates. There is a clear trend of users expressing frustration over the handling of local models and API integrations, suggesting that the transition to the Cortex backend has introduced instability in existing functionalities.
Several issues highlight critical bugs, such as #3410 regarding missing placeholders for empty search results and #3409 about import errors on Windows. Additionally, there are numerous feature requests aimed at improving user experience, including enhancements to model management and UI responsiveness.
Most Recently Created Issues:
Issue #3410:
Issue #3409:
Issue #3408:
Issue #3407:
Issue #3406:
Issue #3405:
Issue #3404:
Issue #3403:
Issue #3396:
This analysis underscores the importance of addressing both critical bugs and user-requested features to stabilize the platform and enhance user satisfaction moving forward.
The dataset contains a comprehensive list of pull requests (PRs) from the Jan repository, showcasing a mix of feature additions, bug fixes, and UI improvements. The PRs indicate active development and community engagement, with notable contributions aimed at enhancing functionality and user experience.
PR #3414: chore: small ui improvement
PR #3413: feat: add blank state component for hub and my model search
PR #3412: fix: avoid show remote model on system monitor
PR #3402: fix: save state toggle input box setting
PR #3354: feat: add tests for migrations
PR #3122: feat: Integrate Devbox shell for virtualized local development
PR #3415: fix: disable remote model when user did not setup apiKey
PR #3411: chore: remove model selection in assistant tab model
PR #3401: feat: enable copy instruction
PR #3400: Chore/simple template report
PR #3399: chore: require yarn 4.3.1
PR #3395: chore: clean up console modal component
The analysis of the pull requests reveals several key themes and trends in the ongoing development of the Jan project:
The sheer volume of open and closed pull requests indicates a highly active development environment within the Jan project. The recent PRs focus on both minor UI improvements and significant feature enhancements, suggesting that developers are responsive to user feedback while also prioritizing stability and performance.
A notable number of PRs are dedicated to enhancing user experience (UX). For instance, PRs like #3413 and #3401 introduce new components and functionalities aimed at making interactions more intuitive for users. This focus on UX is critical for a project that aims to provide an offline AI solution, as it directly impacts user satisfaction and engagement.
The introduction of testing-focused PRs such as #3354 highlights a commitment to maintaining code quality through rigorous testing practices. This is particularly important given the project's complexity and its reliance on various hardware configurations and backends.
The presence of review comments and discussions within PRs illustrates a collaborative culture among contributors. This engagement is vital for fostering a sense of community around the project, which can lead to more robust contributions and quicker resolution of issues.
Several PRs address technical challenges faced during development, such as fixing bugs related to API paths (#3316) or enhancing migration processes (#3354). These challenges reflect the complexities involved in developing an offline AI tool that supports multiple hardware configurations.
The ongoing discussions about integrating tools like Devbox (#3122) suggest that the team is not only focused on current features but is also considering future enhancements that could improve developer workflows and ease contributions from external developers.
In conclusion, the pull requests from the Jan repository reflect a dynamic project with active contributions aimed at enhancing both functionality and user experience while addressing technical challenges head-on. The collaborative spirit evident in the discussions around these PRs bodes well for the project's future growth and stability as it continues to evolve into a robust self-hosted AI solution.
Faisal Amir (urmauur)
Louis (louis-jan)
Van Pham (Van-QA)
NamH (namchuai)
hiento09
Gabrielle Ong (gabrielle-ong)
Mark Nguyen (marknguyen1302)
Overall, the development team is engaged in a dynamic workflow characterized by collaboration, responsiveness to user needs, and a commitment to continuous improvement of the Jan application.