This is offtopic, but I switched from Alpaca to duck.ai. I try not to use AI too often and even though I like the idea to run it locally, duck.ai is way easier to use.
this post was submitted on 28 Jun 2025
83 points (88.1% liked)
Linux
55801 readers
855 users here now
From Wikipedia, the free encyclopedia
Linux is a family of open source Unix-like operating systems based on the Linux kernel, an operating system kernel first released on September 17, 1991 by Linus Torvalds. Linux is typically packaged in a Linux distribution (or distro for short).
Distributions include the Linux kernel and supporting system software and libraries, many of which are provided by the GNU Project. Many Linux distributions use the word "Linux" in their name, but the Free Software Foundation uses the name GNU/Linux to emphasize the importance of GNU software, causing some controversy.
Rules
- Posts must be relevant to operating systems running the Linux kernel. GNU/Linux or otherwise.
- No misinformation
- No NSFW content
- No hate speech, bigotry, etc
Related Communities
Community icon by Alpár-Etele Méder, licensed under CC BY 3.0
founded 6 years ago
MODERATORS
I use it also some times but that is online ai and everything you use it for goes to someones servers so its not private. Alpaca is using ollama to run ai localy on your machine so everything you use it for is private. So those two are completly diferent and are not for comparing which one is easier to use because ease of use is not the point here. Privacy is.
Taking advantage of the fact that this thread became popular, question to all of you guys: do you recommend some other open source LLM front ends?
LM Studio is by far my favorite. Supports all GPUs out of the box on Linux and has tons of options.
LM studio is not open source at all.
Looks like you're right.
I switched to it when Alpaca stopped working on AMD GPUs and was under the impression it is open source.
Depending on how you had it installed, Alpaca split support in the Flatpaks.
If you want AMD support, you need to install
com.jeffser.Alpaca.Plugins.AMD
Doesn't work for me unfortunately, always falls back to CPU ever since the packages were split up.
I was using LibreChat for a while
So far i've really liked just using ollama in the terminal since it just spits out text anyway.
ofc I could even send raw api requests, but sometimes it's good to have a nice GUI that "just works".
Specifically I'm looking for something that could handle not only text responses, but also attachments, speech recognition and MCP support.
Yeah in that case you probably want something else. So far i've only ever used it for text based questions. I think i remember seeing that there is also a webui out there but i don't remember the name.
There are still active accounts on lemm.ee?
I am not certain what you mean by online search function. It can connect to the internet but it doesn't exactly function like a search engine from what I can understand.
Second opcion looks like exactly that..
I used alpaca but they made some changes recently that made it confusing and a pain to use. I deleted it after that as i dont use ai much anyway.
https://github.com/Jeffser/Alpaca
This will probably help anyone unfamiliar with it, since the first search result for Alpaca AI is another online paid AI service which does something entirely different than this. It's used for AI image generation.
The main question I have is since Ollama is optional... If you optionally use it, is it still sharing data with ~~Facebook~~ Meta?
Didnt know that ollama is sharing data with facebook.. Why would it do something like that? Wouldnt that be oposite of what it was created for and that is privacy.. Where did you get that info?
It looked like from comments that's why he made the Ollama integration optional, because some people were concerned since Ollama was built by Meta. It can run without Ollama, it seems.
EDIT: Doing more research on Ollama itself, I'm unconvinced that it's sharing any data, despite being built by Meta.
I didnt know that ollama was built by meta, where did you find that out? Its also an open source project it shouldnt have malicios code like that..
Meta trained and published the model but it’s an open model. I’m not an expert but I don’t believe it’s sharing data with Meta since it’s just the model they trained, you can download it and run it offline. You’re just using the output of all the training they did on your own compute.
You're talking about the llama models, not ollama.
So it doesnt have anything with ollama softvare, you can download any llm it doesnt have to be metas..