Skip to content

Conversation

ServeurpersoCom
Copy link
Collaborator

feat: add dynamic model selector with persistence for llama-swap workflows

  • Add ModelSelector component in ChatSidebar with dropdown interface
  • Implement models store with localStorage persistence and auto-selection
  • Add ModelsService for /v1/models API integration with multi-model setups
  • Inject selected model into chat completion requests
  • Display model capabilities with badges (Vision, Audio, etc.)
  • Auto-refresh server props after model usage for accurate state
  • normalize model display names (base filename + OpenAI-Compat name)

Particularly useful for llama-swap users managing multiple models dynamically
Works with llama-swap workflows or standalone llama-server

@allozaur
Copy link
Collaborator

@ServeurpersoCom thanks a lot for this! I will take a look at this later this week if you don't mind :)

@ServeurpersoCom
Copy link
Collaborator Author

@ServeurpersoCom thanks a lot for this! I will take a look at this later this week if you don't mind :)

SvelteSelectorArrowDOM.hydrationBug.mp4

I've been fighting this hydration bug for 2 hours :( I just want it pixel-perfect.

…flows

- Add ModelSelector component in ChatSidebar with dropdown interface
- Implement models store with localStorage persistence and auto-selection
- Add ModelsService for /v1/models API integration with multi-model setups
- Inject selected model into chat completion requests
- Display model capabilities with badges (Vision, Audio, etc.)
- Auto-refresh server props after model usage for accurate state

Particularly useful for llama-swap users managing multiple models dynamically.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants