The average person wouldn’t be building an open source LLM either

Yeah that’s why I’m saying:

Do you build your own Linux from scratch? If so why would you assume you can build an LLM from scratch?

The OP is basically saying it’s not really open source unless I can personally build it! Which I am saying I don’t think is a requirement of open source software (your personal ability to compile software does not negate from it it’s open sourceness)

tbh I wouldn’t have an idea on how to build either, they are way above my skill level, i have no idea how to make a linux distro either, but i’m certain most are open source

Today, we’re launching Unsloth Studio (Beta): an open-source, no-code web UI for training, running and exporting open models in one unified local interface.

https://unsloth.ai/docs/new/studio

This was only recently released, maybe in the future we’ll have training material uber compressed down in an open source format that anyone with the skill and knowledge can use and different ‘distro’ releases of LLM’s, we already have tons of smaller models especially from European Universities and others

The EuroHPC Joint Undertaking (JU) provides access to the computing time and support services offered by the EuroHPC AI Factories. The AI Factories are open to European users from various sectors, including industry, research, academia and public authorities.

https://digital-strategy.ec.europa.eu/en/policies/ai-factories

We are only like 3-4 years into AI going mainstream if that, afaik the heat death of the universe is at least 1000 years away, we have lots of time to work and improve on them, I can only wonder where they will be at in 100 years, so I try not to make any damning facebook boomer tier statements about the future

It’s a fundamental problem with the fediverse, it’s funny that one of the fediverses biggest features ‘decentralisation’ works against it

Someone had a similar question before about how the place was getting smaller and I actually posted about it somewhere but the original post has been deleted so here’s a repost of my post in response to someone:

You’re 100% right to be concerned and to be honest I have doubts lemmy will ever crack more than a few million users, the same thing happened with Mastodon, something that relies so heavily on volunteers running the infra almost inevitably results in burnout because the fediverse works on a disincentive basis:

Basically the more popular a server is, the more funding it requires, the more admins it requires, the more work it requires, and all of this is on a slim margins or more likely requiring on people to donate time/money/effort ‘for free’ is a huge ask.

The supply of people sitting around doing nothing all day who care enough to dedicate their time/effort/money to running a social network… for free… is a very small group, almost as small as the amount of people who are willing to donate every month to a social network.

You can find mods of communities are usually fans of the communities they mod, it’s a topic they enjoy and so the incentive for them to invest their time is to keep their community clean and great. But running a social network which has hard costs not just time is a whole other thing

This is opposed to a regular website or social media network, where as it gets bigger, it makes more money through ads/subscriptions, the incentive is to get bigger to make more money

And then they can simply pay people to do the shit no one wants to.

The reality for me is that the money has to come from somewhere, you can do a paywall like newspapers do or beg for donations every page visit like the guardian/wikipedia do, or the usual suspect allow advertising, but the money has to come from somewhere.

Thus the fediverse has a disincentive to growing larger, it is simply easier and more sustainable to remain small

So sadly we’ll just have to enjoy our fragmented, over-moderated, over-dramatised, sometimes slow, sometimes down, sometimes goes out of money, sometimes the server owners just burn out, little spot until something better is invented

Because the average person is not building Linux from scratch nor would they know how to

But regardless, the main point of the gap is resources

What makes you think we won’t have the resources in the future?

Any model that can run on 16GB or less, is not going to be any close in real world tasks, to any other cloud based model. It just cannot be.

Well you can compare Gemma 4 running in LM Studio on an average gaming PC to ChatGPT3.5 and you tell me? Or is your benchmark purely based on right at this very moment between open source models today vs cloud today?

For reference Gemma 4 is 26 billion parameters, gp3 thought to be over 175 billion and of course had no optimisations like MoE, it was searching its entire library every single question so was rather slow as well

We know as well that there is no slow down in pushing for optimisations, Deepseeks initial release was the initial driver for you don’t have to just scale up using hardware alone

https://research.google/blog/turboquant-redefining-ai-efficiency-with-extreme-compression/

They’re also pushing with Chinese native chips from Huawei trying to diversify away from nvidia holding the crown

The problem I’ve got is that you all have a god of the gaps, the conversation I was having 3 years ago was different to 2 years ago was different to 1 year ago, I was told AI could never do songs good enough then suddenly people were worried they couldn’t tell the difference, then they said they could never do movies, now apparently not only is it good enough it’s hilarious

https://www.youtube.com/watch?v=fgHn7PI55J4

The open source LLM’s we have today are incredible and in the last few months we’ve had Qwen, GLM, Nemotron/Nvidia, Mistral, Google and heeaaps of others released, it feels like you’re just looking for a reason to be dour and pessimistic but that’s just me

Any way I’m off to sleep, have a good one :)

I have no idea what you’re talking about

When Ubuntu was just appearing I was using Debian. People laughed at me, saying I am a little bit slowpoke

Why are you offended being called a slowpoke using slowpoke OS?

I use it on my server because of this reason

Now it looks like Ubuntu is starting to die

Its been dead according to linux purists since 2011 when they went to Unity over Gnome or KDE so ignore them :P

He’s remembering things from Ubuntu 12.10, yeah 14 years ago :)

yeah mistral is weak like that, I would say they’re at least a year away from chatgpt and claude but they just don’t have the compute

I would recommend to use AI for basic games like translate sentence, anki style flash cards, fill in the sentence, eg. unlike duolingo it won’t annoy the hell out of you with ads and energy

That was a really good post, liked it a lot

A lot more confident we’re not going to end up with AI in my notepad

The model is named Apertus – Latin for “open” – highlighting its distinctive feature: the entire development process, including its architecture, model weights, training data and methods, is openly accessible and fully documented.

https://ethz.ch/en/news-and-events/eth-news/news/2025/09/press-release-apertus-a-fully-open-transparent-multilingual-language-model.html

There is also a move into synthetic data and human trained so we will have to see where the training data goes copyright wise in the future

The useful ones are still provided by big companies because the rest of us can’t afford the hardware to train them.

We have computing power in our pockets a million times more powerful than we used to send man to the moon, why do you think we’ll never have enough power?

I have already pointed out https://eurollm.io/

The EuroLLM project includes Instituto Superior Técnico, the University of Edinburgh, Instituto de Telecomunicações, Université Paris-Saclay, Unbabel, Sorbonne University, Naver Labs, and the University of Amsterdam. Together they created EuroLLM-22B, a multilingual AI model supporting all 24 official EU languages. Developed with support from Horizon Europe, the European Research Council, and EuroHPC, this open-source LLM aims to enhance Europe’s digital sovereignty and foster AI innovation. Trained on the MareNostrum 5 supercomputer, EuroLLM outperforms similar-sized models. It is fully open source and available via Hugging Face.

So long as someone doesn’t want to rely on big tech there will be people pushing for independence just like Linux users such as myself

Do you build your own Linux from scratch? If so why would you assume you can build an LLM from scratch?

Every corporation and government are blindly sticking with Microsoft

Are you sure?

France to remove Windows from government computers in sovereignty push

https://www.rfi.fr/en/france/20260417-france-to-remove-windows-from-government-computers-in-sovereignty-push

https://tuta.com/blog/countries-ditching-microsoft-choosing-linux-digital-sovereignty

I doubt it will happen with governments

It does not take much for things to change, you might like this:

We’ve Hit A Wall With Transport. Here’s Why | Black Swans 3 | If You’re Listening

https://youtu.be/o1R6Aq19A6Y?t=1281

For which you still need massive amounts of memory and compute to run reliably

2026’s average gaming PC is massive amounts of memory and compute apparently

The gap will take decades to close, if it ever does.

lol there are plenty of open source models in the top 100 with multiple SOTA models released in the last few months alone

There’s also smaller LLM’s being made like https://eurollm.io/ which excel in their own ways

That, and the fact that chatbots and agents nowadays rely on all sorts of proprietary customizations

Funny that just came up: https://discourse.ubuntu.com/t/the-future-of-ai-in-ubuntu/81130?=0

Previously, to benefit from the full power of LLMs, you had to skew to higher parameter models. Recent developments in models like Gemma 4 and Qwen-3.6-35B-A3B demonstrate advanced capabilities such as tool-calling which enable LLMs to search the web, interact with external APIs and file systems, troubleshoot live systems and fundamentally reason about topics that lie outside of their initial training data.

The gap will take decades to close, if it ever does.

😁

what a neat little site

Open source models ? https://huggingface.co/ + https://lmstudio.ai/

so the irony is that you guys are hallucinating an answer confidently, AI is really good at translation and language

https://www.deepl.com/en for example is far better than google translate was because they leapt onto ML/AI earlier

If you don’t want to straight up use Mistral/EuroLLM which are trained on European languages:

The company, which launched two years ago, released on Tuesday Mistral Large 3, which the startup claims maintains the same level of performance in “a large number of languages,” particularly European ones.

“Most AI labs focus on their native language, but Mistral Large 3 was trained on a wide variety of languages, making advanced AI useful for billions who speak different native languages,” the company said in a press release.

https://www.euronews.com/next/2025/12/02/mistral-europes-ai-champion-releases-new-smaller-frontier-models-heres-what-to-know

If you don’t want to directly use a chatbot there’s also https://morpheem.org/ and https://languatalk.com/try-langua which uses AI but builds it around the games specifically, vs just straight up telling the AI bot what to do

Hasn’t huge losses to scale up been Silicon Valleys module operadi for a while?

https://www.abc.net.au/news/2019-06-08/uber-netflix-run-at-a-loss-and-they-dont-care-/11185434

i hope people who like the cold are enjoying it atm, as someone who swings towards heat i feeel like if this is autumn then winter is not going to be fun

are you able to point some out on a1/b2 content on mistral for me? im interested to see if you’ve found some