Yeah I enjoy it as well. Just in case you missed it - a fix was merged into llama.cpp two days ago which is said to improve quality.

Edit: I stand corrected - the fix for the issue you’re experiencing has not yet been merged.

Exactly this. Since it does not seem to be federated, you’re still forced to give your data to a third party you can’t choose. And this makes the open source aspect a rather marginal benefit, at least for the privacy-concerned end user. Still, I appreaciate the effort.

I haven’t tried it, but there is one: https://github.com/Alovoa/alovoa

Given that Google generated more than 250 billion U.S. dollars in ad revenue in 2024, I’d say they must be pretty effective.

Source

Depends on the version you’re running.

https://forgejo.org/docs/latest/admin/upgrade/from-gitea/

I see. When I run the inference engine containerized, will the container be able to run its own version of CUDA or use the host’s version?

Thank you for taking the time to respond.

I’ve used vLLM for hosting a smaller model which could fit in two of GPUs, it was very performant especially for multiple requests at the same time. The major drawback for my setup was that it only supports tensor parallelism for 2, 4, 8, etc. GPUs and data paralellism slowed inference down considerably, at least for my cards. exllamav3 is the only engine I’m aware of which support 3-way TP.

But I’m fully with you in that vLLM seems to be the most recommended and battle-tested solution.

I might take a look at how I can safely upgrade the driver until I can afford a fourth card and switch back to vLLM.

I use the the proprietary ones from Nvidia, they’re at 535 on oldstable IIRC but there are a lot newer ones.

I use 3xRTX2000e Ada. It’s a rather new, quite power efficient GPU manufactured by PNY.

As inference engine I use exllamav3 with tabbyAPI. I like it very much because it supports 3-way tensor paralellism, making it a lot faster for me than llamacpp.

I use the the proprietary ones from Nvidia, they’re at 535 on oldstable IIRC but there are a lot newer ones.

That brian typo really gave me a chuckle. Hope you found the movie you were looking for.

Wikipedia states the UI layer is propriertary, is that true?

The country’s official app for COVID immunity certificates or whatever they were called was available on F-Droid at the time.

Too bad they’ve only been dropping dense models recently. Also kind of interesting since with Mixtral back in the days they were way ahead of time.

A review from earlier this year didn’t sound too bad.

Edit: as pointed out, the review seems to be about the previous version of the phone.

I’d add that memory bandwidth is still a relevant factor, so the faster the RAM the faster the inference will be. I think this model would be a perfect fit for the Strix Halo or a >= 64GB Apple Silicon machine, when aiming for CPU-only inference. But mind that llamacpp does not yet support the qwen3-next architecture.

One reason could be that the audience on lemmy has a left-ish bias and there’s a political component to the Spotify exodus.

Edit: don’t get me wrong, I love seeing content and engagement on here.

SFTPGo is such an awesome project, never had any problems with it.

Some people on another discussion platform were praising the new Mistral Small models for agentic use. I wasn’t able to try them myself yet, but with 24b params you would certainly fit a quantized version in your 24GB.

Thanks for the tip about kobold, didn’t know about that.

And yeah, I can understand that building your own rig might feel overwhelming at first, but there’s tons of information online that I’m sure will help you get there!

Alright thanks! I found it somewhat difficult to find information about the hardware requirements online, but yeah, maybe I just have to try it.