Searching for models returns f32 models and no option to select the quant size

[sammcj]

Bug Report

Searching for models gives no option to select the quant size and seems to favour f32 which is a bit overkill IMO 🤣, searching with a specific quant size returns a completely different model

Steps to Reproduce

1. Search for Phi 3.1
2. Returns results for F32
3. No option to select the quant type/size
4. Search for Phi 3.1 Q6_K
5. Note that it returns results for llama3?

Expected Behaviour

Be able to search for a model and have it suggest a reasonable quant size (e.g. Q6_K, Q5_K_M and Q4_K_M) by default.

Be able to search for a model and it's quant size and have it return that model and quant size (rather than a different model entirely)

Your Environment

• GPT4All version: 3.0.0
• Operating System: macOS (latest)
• Chat model used (if applicable): N/A

[uninstall-your-browser]

It would be nicer if it just listed the quant options available like lm-studio does