FEATURE: A "load" command for loading the embedding model

[Propheticus]

I've made a quick bat file to automatically start LMStudio server + Load the model and then start AnythingLLM. This works, with one caveat; I can't start the embedding model in LMStudio this way.

@echo off
::Starting LMS local API server
lms server start --cors
::Loading a model
lms load Meta-Llama-3-8B-Instruct-Q8_0.gguf --gpu max --identifier Llama-3-8B-Q8-8K
::Start Anything LLM
cmd /c start "" "%localappdata%\Programs\anythingllm-desktop\AnythingLLM.exe"
exit

Would be nice if a next version of lms could be used to load that as well.

[Propheticus]

For reference, current result is that the list command sees the model, but the load command cannot find it:

[ryan-the-crayon]

We currently do not support loading/unloading non-LLM's with lms. We will for sure add support for it in the future. I will update here once it is supported.

[davidmoore-io]

+1 on this.

1. lms list shows all my downloaded llm's and embedding models
2. lms load sergkisel3v/LLama3-8b-Instruct-GGUF-fp16 loads fine, as expected, which is awesome and I'm so glad you guys built this thing.
3. lms load CompendiumLabs/bge-large-en-v1.5-gguf fails, saying the model can't be found despite it being included in the lms ls list.

I'm totally happy to have a dig into it and help build the functionality if nobody's working on it @ryan-the-crayon . I had quick poke around https://www.npmjs.com/search?q=lmstudio and it looks like you own it all, so I figured you're the person to ask?

[DrewLeetw]

+1 on this. lms CLI can only load gguf model, but if I want to load embedding model, now I have to open GUI and click on it. Thanks!

[KungFuFurniture]

+1