jwitthuhn

For immediate revenue yes, but if you want advertisers to keep coming back you need to give them a good conversion rate.

It is very fortunate for us that the authors were kind enough to demonstrate this has nothing to do with safety by adding this exemption.

"Renting attack capacity from [cloudflare]" is inaccurate as I understand things. That group hosts their site behind cloudflare but I have not seen anyone claim that cloudflare's infra is used for the attacks.

This whole article seems conflate hosting an informational site run by the attackers and hosting the attack itself.

I need scissors, 61!

Could you cite the specific law that makes it illegal for someone to export their thoughts?

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.

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.

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'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.

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.

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

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.

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

PVS by my understanding will only ever over-count visible polys.

It is essentially a set of all polygons that are visible from any point inside a fixed volume, but the camera only exists at a single point inside that volume so there will probably be some polys that the camera has no LOS (though I suspect these would still be 'rendered') to and a bunch that are out of the view frustum which will not be rendered.

edit: To observe this you can also load any HL1 engine game, run `r_speeds 1` in the console, then it will show you how many world polys are currently being drawn in the corner of your screen, which is probably the count referenced by John Romero here.

The most recent integration I've seen is is OpenMW, which is an open source re-implementation of the Morrowind game engine. Basically it is built on the assumption that people are going to make mods that do a ridiculous amount of number-crunching in lua so any small improvement to performance is welcome.

At 4 bits that model won't fit into 128GB so you're spilling over into swap which kills performance. I've gotten great results out of glm-4.5-air which is 4.5 distilled down to 110B params which can fit nicely at 8 bits or maybe 6 if you want a little more ram left over.

If you have 20 hours to spare I highly recommend this youtube playlist from Andrej Karpathy https://www.youtube.com/watch?v=VMj-3S1tku0&list=PLAqhIrjkxb...

It starts with the fundamentals of how backpropagation works then advances to building a few simple models and ends with building a GPT-2 clone. It won't taech you everything about AI models but it gives you a solid foundation for branching out.

Rosetta 2 is going to be EOL'd within the next few years. A more permanent solution would certainly be welcome.

You are describing Windows 11 LTSC which is a product that exists because Microsoft knows people want to turn this crap off.

It is of course only available in volume licensing to keep it away from normal users. Only businesses get to control their computers.

Yes the security of every Windows computer was much better then, any software that automatically updates itself without user consent is obviously a massive security risk because the user is no longer in control of what software they run.

I would think it is pretty silly if I needed some sort of verification to drive people I personally know around because other people were getting their car hijacked after choosing to pick up strangers they found on the highway.

I see them going for $150-300 on ebay, just don't ask where they came from.

It is wrong to release something unreliable even while acknowledging it is unreliable? The product performs as advertised. If people want accurate information an LLM is the wrong tool for the job.

From the Gemma 3 readme on huggingface: "Models generate responses based on information they learned from their training datasets, but they are not knowledge bases. They may generate incorrect or outdated factual statements."

So you can only use it if Canva has your email and they can send you exciting new offers.

Worst part about this is that these days the error message simply lies to the user. "Whatever.App is damaged and can't be opened. You should move it to the trash."

If you remove the 'quarantine' attribute that gets added to downloaded files it runs great.

Somewhat related: I wrote up a MTG card generator based on nanoGPT a while ago that I think produces pretty good results for being 1m parameters.

The real neat thing about this is that WotC makes a few thousand new cards each year, so my training data set just grows over time and the model gets better with no effort spent on my part.

https://github.com/jlwitthuhn/TCGGPT

If they are worried about the cost of initial ingestion then a gate on enabling would make a whole lot more sense than a gate on disabling.