🐛 [Bug]: Minotaur-15B model stuck at download and does not run about serge HOT 4 CLOSED

@pabl-o-ce I dont remember why those got removed, you remember?

@CorentinWicht Looking at StarcoderPlus readme:

Please note that these GGMLs are not compatible with llama.cpp, or currently with text-generation-webui.

So we can't use it

from serge.

@gaby yes this are GPT2 architecture models so python-llama-cpp only work exclusively for GGML and now GGUF (it was my mistake to added)

from serge.

@CorentinWicht The list on the readme is the one from main. If you are using latest or a tagged release that may be why of the mismatch.

I plan on doing a release, but there's some bugs that need to be fixed first.

Try using the main tag, and see if the problem persists

from serge.

@CorentinWicht The list on the readme is the one from main. If you are using latest or a tagged release that may be why of the mismatch.

I plan on doing a release, but there's some bugs that need to be fixed first.

Try using the main tag, and see if the problem persists

Very good, thanks for your prompt reply.
But why where models like Minotaur-15B introduced in v.0.3.0 and removed right away? #541

Any chance a model like TheBloke/starcoderplus-GGML be supported in upcoming releases?

from serge.