The checkboxes inform the model as well as the user, and you can observe this yourself. For example in a C++ project with MyClass defined in MyClass.cpp/h:

I ask the model to rename MyClass to MyNewClass. It will generate a checklist like:

- Rename references in all source files

- Rename source/header files

- Update build files to point at new source files

Then it will do those things in that order.

Now you can re-run it but inject the start of the model's response with the order changed in that list. It will follow the new order. The list plainly provides real information that influences future predictions and isn't just a facade for the user.

And when it doesn't, it politely apologizes, at least :)

If the LLMs run by these people are turning up real bugs then their confidence in touching kernel code seems pretty earned, imo.

I've been using qwen-code (the software, not to be confused with Qwen Code the service or Qwen Coder the model) which is a fork of gemini-cli and the tool use with Qwen models at least has been great.

My favorite as a kid was also in a Zelda game.

In the original (and maybe also DX) release of Link's Awakening, the game uses a top-down view with the world split up into tiles. Walking of the left side of a screen makes you end up on the right side of the next screen over.

What you could do is pause at the right frame on the screen transition, and you would end up on the new screen but link's position would not change. So you walk off the left side of a screen and end up on the left side of the new screen. Lots of fun to be had with skipping important stuff with that.

I've been largely using Qwen3.5-122b at 6 bit quant locally for some c++/go/python dev lately because it is quite capable as long as I can give it pretty specific asks within the codebase and it will produce code that needs minimal massaging to fit into the project.

I do have a $20 claude sub I can fall back to for anything qwen struggles with, but with 3.5 I have been very pleased with the results.

How much VRAM do you need for that?

Not OP, but I ran 122b successfully with normal RAM offloading. You dont need all that much VRAM, which is super expensive. I used 96gb ram + 16gb vram gpu. But it's not very fast in that setup, maybe 15 token per second. Still, you can give it a task and come back later and its done. (Disclaimer: I build that PC before stuff got expensive)

128GB on a mac with unified memory. The model itself takes something like 110 of that and then I have ~16 left over to hold a reasonably sized context and 2 for the OS.

I do have a dedicated machine for it though because I can't run an IDE at the same time as that model.

I squeeze Qwen3.5-122B-A10B at Q6 into 128GB. It's a great model.

Wow what kind of hardware do you have? Mac Studio, dgx spark, strix halo? How fast is it?

Strix Halo, I'm seeing performance inline with these results[0].

I'm interested to investigate the claimed gains from the lemonade-sdk port of Apple MLX inference[1].

[0]https://kyuz0.github.io/amd-strix-halo-toolboxes/

[1]https://github.com/lemonade-sdk/lemonade/issues/1642

It only potentially saves money for people on API pricing, it exhausts tokens faster with no benefit for users on the Claude Code subscription. Those users had their cache TTL reduced from 1 hour to 5 minutes and are saving no money because they were not paying based on the cache time in the first place.

So they were not "pressured" but Atari contacted them and they proceeded to make this decision based because they "needed to balance Atari’s commercial interests".

That sound indistinguishable from being pressured.

I think they're saying Atari didn't threaten them but they both understood that they could have. Honestly it sounds like Atari were trying to be nice. Like "you technically aren't allowed to do that, and we could just set our lawyers on you, but we'd like to not do that while also making money on our re-release".

This seems like a perfectly reasonable compromise to me.

How is "I haven't talked to my lawyer yet but you know I could" not a threat/pressure?

In the same way that "you kids aren't allowed to skate here, but maybe if you do it over there I could just turn around and not notice you" isn't a threat.

"you kids can't skate here" is the threat part.

That's just stretching the definition of "threat" beyond its normal meaning.

If I tell my kids "it's bedtime" is that a threat?

If they say "no" are you gonna let them stay up?

If not then it may not be a literal threat but it contains the implied threat

There is no reason to assume they said that and all the reason to assume they didn't say that.

Reaching compromises with others is part of life. If the question is whether a copyright from 1995 should hold in this case, I would say no. But the world is sometimes not as we may want it to be. So taking that for granted, this seems like a very reasonable and mature resolution.

Indeed. It sounds like they were further pressured to say they were not being pressured.

The types of folks who make reimplemented game engines often do it as a labor of love towards the original. And the best companies often have great appreciation for their modding communities and preservationists. (Witness the good collaborations between some companies and SCUMMVM, for instance.) This may well have been a conversation that was entirely reasonable and respectful.

I just can't believe that given the outcome and the wording of the posts from the project. If there was respect here there would have been no threats. If there were no threats there would be no talk of "balancing commercial interests"

Why should an OS demand personal information from its users? It creates an unnecessary risk that the information will be leaked.

Laws exist that dictate what apps are allowed to do depending on the user's age. This means that in order to follow the law they must collect the user's age. If collecting the user's age is a common requirement of apps it makes sense for the operating system to expose an easy way to do that to make app development easier on that platform.

No, it makes sense for an App Store to do that. Or, that HTTP headers are set at the device or network proxy.

User account creation wizards could just create the dot files for the App Store. These weird laws ban OS.

The issue is that software can be installed from outside of the app store on pretty much every OS. Also if you have multiple app stores it would be convenient if they could all get it from the same place.

I'm skeptical we should ban all operating systems which permit this without an interactive age check. Shouldn't the free market acolytes be arguing that parents can choose between competitors which offer ever-improving parental controls?

I agree with you, but it seems like the free market is moving too slowly in regards to this pain point for parents which is resulting in demand for them to force change through the legal system.

In the US there is a certain class of politician that considers poor people being able to exercise their rights a problem that needs to be solved.

Is that really limited to the US though?

The important distinction here is that CD-ROMs can store data indefinitely, but CD-Rs and CD-RWs can not.

BD-R and M-Disc CD and DVDs have archival storage life.