ed_mercer

When I saw 2 vCPU and 4GB, and then below that "4 GB persistent disk" I was like ooohhh it's the disk, not the memory. But it IS actually 4GB memory + 4GB disk.

This was so fulfilling to do on x.com

jev is optimized for it, a standard LLM isn't and it also costlier and slower.

So are they going to release this tuned model? Or keep it for themselves?

I also noticed this personally. Consider when an agent says

"Should I start with the spec, or go straight to building it?

Unless you're working on a super crucial piece of engineering, you can probably get away with going straight to building it. Even if something doesn't go the way you intended, I find that it's usually faster to correct the agent later, once the initial implementation is in place. It's more of an iterative approach to building and I feel that it is less cognitively demanding.

Except for VRR, HDR, touch id, video encoding and a sleep mode that doesn't drain your battery.

Does this extend to open models like GLM 5.3? This would mean that simply changing the harness to Pi reduces cost in half?

Lots of criticism here but I think this is extremely promising. When this tech is extended to videos and perhaps even other forms of media, I think it has the potential of stopping all slop!

email? Is a user's email up for grabs just like that?

>I do enjoy using their free AI

It's really not free. You're paying with your data.

you mean something like step level debugging for agents?

Meanwhile, 7 years on my F-91W

Why would you ever connect a TV to the internet in the first place?

Awesome write-up, thank you! This makes me want to try this out as well.

Hi, can you elaborate on this? Why not use Asahi Linux?

Fable-class models IMO are now capable enough to maintain your code as well.

This is all great until you change or upgrade to a newer model, and you can start over again because some of the directions/rules are now part of the model.

OP was doing 30K USD MRR, not sure why he needs to find a job at all. Did he spend it all?

> which has even seen Google co-founder Sergey Brin return to day-to-day operations

This is false, Brin already returned to active work at Google in early 2023. https://www.forbes.com/sites/richardnieva/2023/01/31/sergey-...

I personally got used to the claudisms and think people are making too big of a deal out of it. Yes it says load-bearing and caveat a lot, but at least you know behind the scenes what meaning it's assigning to certain situations.

getting a 502

It's the predecessor. See https://news.ycombinator.com/item?id=38415092

Alex's first post about OpenRouter got 6 upvotes and 0 comments on HN https://news.ycombinator.com/item?id=35481760

I'd be curious to see how this would run on Qwen3.8 27B

Yes, he has financial incentives. That doesn't automatically make what he says false or dishonest.

I agree, and I'm actually pro Opus 5 exactly for this reason. In my opinion, agents talking to agents is the future, and humans will move to a higher abstraction layer. So it's the right move to make for Anthropic.

Many of the complaints that people are having with Opus 5 are actually acknowledged and explained on opus's 5 prompting guides (https://platform.claude.com/docs/en/build-with-claude/prompt...)

It also seems naive that Anthropic, with some of the smartest people on the planet working on AI, would not know about this.

If it can save 10% of tokens, how is that useless?

There is no positive future with meta in it.

They're pointing to macOS and saying it uses 5 times less RAM there, but is 230MB for a weather app really something to be proud off?