jwitthuhn

Not really, on the project I'm working on right now most of my longer prompts seems to consume somewhere between 300k and 400k tokens (input + output) and I find that both Opus 4.8 and Fable 5 don't seem to get lost with that much context. At least the results still look good to me.

I don't think I've ever pushed past 500k though, it may well fall apart at the higher end of what is officially supported.

Claude does the same when you hit the context limit, but the difference is that claude's context limit is 1m and Codex's is 272k now. Claude lets you get a lot more done before you have to worry about the start getting compacted away.

The text you linked mandates, as the first technical requirement, "An ADDW system shall determine when the driver’s visual attention is not directed towards the driving tasks and alert the driver through the vehicle human–machine interface."

Can you describe how you believe the driver's visual attention can be tracked by anything other than a camera pointed at the driver's face?

If no such other system exists, how is this doing anything other than mandating cameras pointed at the driver?

Why would physical copies matter for this? All physical games have to be signed by Sony anyways so it's not like a third-party can produce them.

This is true but the buyout happened in 1985. Servco Pacific has been in charge for longer than Leo Fender and CBS, the two previous owners, put together.

They did upload the wrong model but as of the time of writing they have not fixed it. Right now, 12 hours after they took the old one down, there is simply no model present in their huggingface repo.

Fortunately for us all Dario literally asked for this sort of restriction on model usage so we can be certain Anthropic are not victims here.

Same for switch 1/2. it even has a browser built-in so you can log in to wifi captive portals but you can't access it through any normal means.

It is not a backdoor if I authorize some other app to use that data, it is a front door.

Apple thinks users should not be able to make the decision about who can access their data. It is not more complicated that that.

Thanks for the correction, I had misunderstood that and thought it applied more broadly but it is indeed just "source code".

Given that, I'm glad to see something that protects software more broadly.

Good, Bernstein v. United States already established that software is speech. Limitations on what software one is allowed to produce are very blatant prior restraint.

And occasionally un-releasing them like with WizardLM.

Forks are still alive on github, so it seems unlikely microsoft did this to suppress the code. Unless they are wildly incompetent, which I don't want to outright reject as a possibility.

https://github.com/xiaoji235/bitlocker-bypass-tool-for-winre

Unfortunately I don't think there is any way to see a list of all the forks now that the main repo is dead, but you can search the phrase "A huge thanks to MORSE, MSTIC and Microsoft GHOST for making this public disclosure possible" to find more copies.

Yeah I've also found that models are good enough that the extra spend on premium models isn't always worth it, particularly for my small personal toy projects.

A $20 claude sub goes a long way when you plan with Opus and execute with Sonnet.

"One can point out that such electoral pitches are dishonest bullshit until one is blue in the face"

This is what I intend to do. To just let lawmakers get away with passing bullshit because "of course they will" is a defeatist attitude that I don't advocate.

That is why it is so important to be loud about this incredibly obvious disingenuous behavior. Make sure everyone knows the people advocating this do not care one bit about children's safety.

It is an admission from the writers that this law is unrelated to safety and people should very loudly and frequently point that out.

If OSes that don't verify the age of their users are a genuinely unsafe for children, why should they be allowed just because they are open source? That doesn't seem to mitigate dangers associated with age in any away I can identify.

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.