KerrAvon

what was the problem with 5? to me, it seemed like the first Opus since 4.6 that was a real step up in intelligence without any obvious downsides

Wikimedia's entry:

https://give.org/charity-reviews/other-charitable-organizati...

literally says it meets standards and all the items green? is there a leaderboard somewhere you're referring to?

edit: s/the board/all the items/ for clarity

DS 4.1 Flash absolutely has guardrails. Maybe not ones as effective as some models.

>They will after a few incidents where unguardrailed local models are used to hack and stop the water supply.

This is irrational AI fearmongering. Please read https://sharptext.net/2026/some-of-all-fears/

> Market places where drugs/guns are sold are not, they are all infiltrated and shut down after a while.

might want to look up "gun show loophole"

There are pictures, dude. What is wrong with you?

Haiku should rethink that AI rule. Put very high code quality gates in place, sure, but Haiku in particular is a very low-velocity project and AI could help them so much. There is no reason you can't produce high-quality code with AI conformant to your style guidelines and structural preferences, it just takes a lot more effort than the slop.

I think it'll need a compelling DAW to make that a reality.

Not entirely sure what point you're trying to make, so maybe this is not relevant to it, but, in 2026, satellite emergency support makes it seem irresponsible to not take your iPhone if you value your life or have other people in your life who would miss you if you were gone.

100%. LGs work perfectly well in this role.

FWIW, I'd bet you'd never have noticed the difference.

it is once again time to balance out paulg sycophancy https://idlewords.com/2005/04/dabblers_and_blowhards.htm

top comment right now on this very page: "But actual results inside phone with smartphone cooling and wattage requirement became close to 3300"

What's amazing is that all of these are true at once. If you allow for some significant slack in what "superintelligence" means.

> the 24GB card that's been sitting in gaming PCs since 2022.

I'd wager most people have less. In 2022 a 3080 might have 12 GB if you were lucky, 10 if you weren't -- and you paid for the privilege. A current RTX 5080 is only 16GB.

> Claude deserves respectful engagement and needn't apologize when the person is unnecessarily rude: accountability without self-abasement, excessive apology, self-critique, or surrender. If the person becomes abusive, Claude doesn't become increasingly submissive. The goal is steady, honest helpfulness: acknowledge what went wrong, stay on the problem, maintain self-respect.

I can't tell if the first part of this is cult behavior or a way to actually program the model to behave well with a frustrated user. Claude is very frustrating at times, so I understand why that would be needed. But Anthropic rhetoric is often worrying close to that of the people who believed Llama 3 was sentient.

If I'm doing the math right, that's like 17 t/s? I haven't played with Qwen 3.8 yet, but that seems really slow for a 27B on an M5 Max with sufficient RAM to hold it in memory.

So have you looked at what's happened in the US over the past 10 years?

The US has much further to fall, but it's falling very, very quickly and if there's ever another Democratic president they're going to have to rebuild a lot of the government from scratch.

Tesla has lost both house battery and car sales in my family -- we're talking hundreds of thousands of dollars -- simply because we don't trust him not to remotely shut off our power/cars for petty political reasons.

Also, if you want true privacy you should run AI models on local hardware. (Guess which country's models dominate SOTA/near SOTA open weights? Yes, it's China, and it's not even close. You can run full-fat DeepSeek locally for (just) under $10K USD.)

Funny, the Germans said the same thing a long time ago!

This was the basic theory behind Tesla eventually being able to stop killing Tesla drivers with FSD. It has not worked out.

They don't control that. Pricing is controlled by the question of when RAM is once again made of semiconductors instead of unobtainium.

this is not how it's going to go if OpenAI and Anthropic get their way and the US outlaws use of open-weight models

The US is still fixable if young people will just fucking vote. Irreparable damage has been done, which means the old structures are gone, but they can be replaced with new ones -- a hypothetical Democratic president would need to clean house top to bottom. Obama's greatest flaw was that he didn't see the need to do that after Shrub.

You're making assumptions that aren't supported if you read the actual lawsuit.

We're on the cusp of Kimi K3 becoming usable on sub-10k hardware.

https://github.com/gavamedia/deltafin

128 isn't enough, but if this report is accurate, the next Mac Studio could run it

https://www.macrumors.com/2026/06/25/m5-ultra-mac-studio-202...

IME: Gemma is not great for programming, but it is fantastic at following directions compared to anything else in its size class.

"Better things aren't possible" is a terrible outlook. There have been real improvements in this space, such as passkeys, and recognition that some of this stuff, like frequent password changes, is counterproductive.

It's actually a little harder than people outside the UI framework space might think (there's really not a lot of time per frame; it's easy to put too much work where it doesn't belong; it's easy to regress if you're not monitoring for it after code changes).

It's core to usability at this point, though. They should make the effort.

It's relevant politically, and the discussion in question is political.