LM-Studio with Radeon 9070 XT? Im upgrading my 10GB RTX 3080 to a Radeon 9070 XT 16GB this week and i want to keep using Gemma 3 Abliterated with LM Studio. Are there any users here who have experience with using AMD cards for AI? What do i need to do to get it working and how well does it work/perform?

I've heard of LM Studio being recommended, but usually people discount it due to it not being open-sourced. From what I could see, it's available in code on github, so isn't it open-sourced? Is it that there's a website version they host, which people think is the only one usable?

Trying to find an uncensored model to use in LM Studio or anything else really to get away from the god-awful censoring were seeing in mainstream models.

I have been using both Ollama and LM Studio for a while now. Is there any way to use the models downloaded using Ollama in LM Studio (or vice-versa)?

LM studio doesn't have support for directly importing the cards/files so you have to do it by hand, or go download a frontend like sillytavern to do it for you.

There's definitely something wrong with LM Studio. I've tested it against Ollama using OpenWebUI using the same models. It's dogshit slow compared to Ollama.

LM Studio high CPU usage on Windows I just downloaded the latest LM Studio 0.2.10 and llava v1.5 13B in gguf format to try to do some image interrogation. When I'm trying to interact with the model, my CPU usage goes over the roof because of the WMI provider host process.

It appears that running the local models on personal computers is fully private and they cannot connect to Internet. Can someone please enlighten me on the privacy part just to be sure that I can trust putting personal/work information, project ideas, etc in the chats? Thanks for your help Share Sort by: Best Open comment sort options Best Top New Controversial Old Q&A Add a Comment ...

I dont think LM Studio has extensions. oobabooga is bar the best, most features in so many different aspects. My dude just dropped a new update video on setting it up yesterday.

LM Studio is closed source, so there is no way for me to submit a pull request or make sampler changes to it like how I could for llama.cpp. Those who use LM Studio will have to wait on the developer to implement it. Anyways, I hope this post helps people figure out questions like, "why does this preset work better for me?"