The load-bearing seam stays, not taking that away;

> Every single model is (and always was) behind anything from China.

My guess is, that Mistral plays fair and doesn't try to extract data from US models, while Chinese companies are not.

https://github.com/apple/container/blob/main/docs/technical-... read documentation mr. rare breed

Like, this doesn’t answer when to use this vs Docker. Any reference there?

This is not the apple container framework / that’s the docker alternative. This is a project to easily run full Linux machines in macOS containers.

Since when are you a moderator?

We're all a part of deciding what culture we want to have in our communities. Culture is what we make it, and I don't want a space where people use that sort of language regardless of whether anyone is a mod or not.

Since when are you a meta-moderator? ;)

Why do you think they acquired Bun. To look at it?

Also you needed a leak to know that CC, that is out for 1 year, is in TS?

> Why do you think they acquired Bun

I don't know why they acquired Bun. Surely not because there's no other means to write command line programs.

> Also you needed a leak to know that CC, that is out for 1 year, is in TS?

What, are you shaming people for not being aware what (obfuscated) code some 1 year old software is written in?

So what do you think the obfuscated .js code in a 2026 engineering project was generated with? Nim?

When you use a calculator, did you really solve it or was it the calculator?

With a calculator I supply the arithmetic. It just executes it with no reasoning so im the solver. I can do the same with an LLM and still be the solver as long as it just follows my direction. Or I can give it a problem and let it reason and generate the arithmetic itself, in which case the LLM is effectively the solver. Thats why saying "I've solved X using only GPT" is ambiguous.

But thanks for the downvote in addition to your useless comment.

bruh it’s like Wordle, come back tomorrow

I have had great experience scripting and running http://mitmproxy.org for these purposes. I also have set it in production as a dumb caching proxy for upstream services (We do a lot dumb GETs to list/enumerate)

you run FEX, not direct ARM games

That doesn't magically fix the Qualcomm GPU or the drivers.

The GPU is fine and the drivers Valve are using, if their past hardware is any indication, will be open source. Doesn't magically fix them, but it does allow for Valve to fix them.

If valve can convince Qualcomm to open source their GPU drivers I will eat a banana peel. They would need to write a new one from scratch (or reuse freedreno, but those are going to have performance issues).

Not the first time Valve funded the development of FOSS drivers. They've already done so with Intel's Vulkan stack on Linux, AMD (AMDGPU) and Nvidia (NVK).

SteamOS's core functionality leans heavily on Mesa and there's been a lot of commits for the Adreno 750 lately, mostly coming from Linaro.

Fair enough. The problem is usually in terms of getting enough documentation and help to do this in a cost effective manner. If they manage to pull it off, I will start to second guess every phone OEM who seems to not go down this road.

I don’t think you will be on latest nightly. LTS are good and stable, if FEX is targeting those specs I don’t see a stability issue.

It kinda does. Qualcomm's DirectX drivers were the big issue, and Valve is using Mesa instead.

I have 3 Nikon Coolscans and I also repair them…

I would love a new scanner for 21st century but there just no way anyone serious is trading CCD (or PMT if you got the cash) for CMOS.

But I applaud the initiative and will definitely buy it to try but not to keep.