Skip to content

Conversation

@haddadrm
Copy link
Contributor

@haddadrm haddadrm commented Apr 9, 2025

This PR adds several improvements to Perplexica:

  1. LM Studio providers
  2. Enhanced the Library interface with search and batch delete functionality
  3. Improved and optimized the Discover section with personalization features
  4. Implemented provider name formatting for better UI display

haddadrm added 6 commits April 9, 2025 14:06
…ded category filters and personalization features - Implemented support for multiple languages and search engines - Added preferences saving - Fixed Chinese language search issue
…r to support OpenAI compatible API - Implemented chat and embeddings model loading - Updated config to include LM Studio API endpoint
…ibility

- Add PROVIDER_INFO metadata to each provider file with proper display names
- Create centralized PROVIDER_METADATA in index.ts for consistent reference
- Update settings UI to use provider metadata for display names
- Fix client/server compatibility for Node.js modules in config.ts
@ItzCrazyKns
Copy link
Owner

So from my testing, the discover route has become quite slow. I don't think we need to gather articles on language basis (maintaining and putting a static list with that many values isn't something I am quite fond of)

@haddadrm
Copy link
Contributor Author

haddadrm commented Apr 11, 2025

So from my testing, the discover route has become quite slow. I don't think we need to gather articles on language basis (maintaining and putting a static list with that many values isn't something I am quite fond of)

Not an issue, I split this into 3 PRs:

  1. LM Studio Provider PR (Feature/LM Studio provider and Provider Formatting #743):
  1. Library Improvements PR (Feature/Library Improvements #744): 1e87175 & 6f69f6b (Library enhancements)

  2. Discover Features PR: (hold off on) 5138182 for now to address your concerns

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants