jan
jan copied to clipboard
epic: Users can manage local API server via Dashboard
Motivation
- Users sometimes just want a local, OpenAI compatible endpoint, for other app integrations
- Users want to see server logs, change server settings while using this local server
Specs
- Users can turn on/off servermode, and configure various server settings
- Users are informed through UI, that ServerMode disables other UI level features like Threads
- Users can see server logs
- Users can quickly get linked OUT & access docs
Designs
Preview:
Tasklist
- [ ] #1251
Not in Scope
- Users can edit model settings in servermode
- Postman like playground for users to construct requests
- Embedding api docs within the dashboard (UX antipattern)
Appendix
Previous context:
Model Selection:
- As a user, I want to choose which model I want to make available at https;//localhost:1337 as an API
Server Configuration Options:
- As a user, I want to customize the server port to ensure compatibility with my system's setup.
- As a user, I want to manage Cross-Origin-Resource-Sharing (CORS) settings to control access permissions for security reasons.
- As a user, I want the option to enable or disable verbose server logs for effective debugging and troubleshooting.
Examples (Request, Response, etc.):
- As a user, I want access to clear examples of API requests and responses for better understanding and usage guidance; and if possible, a link to Jan's API docs for more details.
Server Logs:
- As a user, I want access to detailed and comprehensive server logs, providing in-depth information about each API request and response for thorough debugging and analysis.
- As a user, I want a dedicated section displaying API metrics, offering insights into usage statistics, response times, and other relevant performance indicators for better optimization and monitoring.