feature: model gallery

[mudler]

A curated gallery of models:

• support for multiple repositories (so anyone can build their own personal gallery)
• automatically download, setup model with default config files
• feat: add /models/apply endpoint to prepare models #286
• automatically load it in memory so it's already ready for inference
• Wizard in the index.html -> if no models, propose to install a new model
• UI to select and install models from repositories

[localai-bot]

⚠️⚠️⚠️⚠️⚠️

Hi! I'm a bot running with LocalAI ( a crazy experiment of @mudler ) - please beware that I might hallucinate sometimes!

but.... I can also be funny or helpful 😸 and I can provide generally speaking good tips or places where to look after in the documentation or in the code based on what you wrote in the issue.

Don't engage in conversation with me,
I don't support (yet) replying!

⚠️⚠️⚠️⚠️⚠️

To add support for multiple repositories, we can create a new API endpoint that allows users to add their own repositories and manage them. This endpoint can accept a JSON payload with the name of the repository, its URL, and other metadata such as description and license. Once added, users can then install models from these repositories using the /models/apply endpoint as usual.
To automatically download and setup model with default config files, we can create a new API endpoint that accepts a YAML file with the model configuration. This endpoint will then download the model files from the specified URL and save them in the models directory. It will also generate default config files based on the model's metadata and save them in the conf directory. Finally, it will load the model in memory so it's ready for inference.
To automatically load it in memory so it's already ready for inference, we can modify the /models/apply endpoint to load the model in memory when it's installed. This way, users won't have to manually call the /models/load endpoint to load the model after installation.
For the Wizard in the index.html, we can add a button that checks if there are any models installed and proposes to install a new model from the gallery if none are found. We can also add a list of recommended models to get started with.
For the UI to select and install models from repositories, we can create a new page in the dashboard that lists all the available repositories and their models. Users can then select a model and click on it to install it using the /models/apply endpoint. We can also provide search filters to help users find specific models based on keywords or other metadata.

Sources:

• https://github.com/go-skynet/LocalAI/tree/master/pkg/gallery/request.go
• https://localai.io/models/index.html

[mudler]

most of it is done - the webui is not there but it is tracked by several other issues e.g #918