dist-epoch

and zstandard outdoes 7zip (LZMA)

Interesting.

I'm on the other side, I hate webp because if I download such an image a lot of apps don't support it. So typically I need to screenshot it instead to get a usable jpg/png.

Now one more format to hate.

I would say webp/jxl/avif is anti-general compute, since they require very recent software.

A phone is a much more secure environment to run untrusted apps than Linux/mac/Windows since the apps are sandboxed.

On Linux is free for all, a regular app running as a user can read/exfiltrate anything owned by that user.

And sandboxing like snap/flatpak is not popular.

Since you're already setup, try adding padding to the input prompt. Add 1000 dots "." after the question, this provides extra space where to reason about the problem. It's not as good as real reasoning, since it's not recursive, but it does improve the quality.

https://arxiv.org/abs/2510.01238

One of the largest group of people which benefit from AI are programmers.

Do you think they also are mostly deficient of scruples?

Context is one thing, but this article implies it goes into psychological analysis territory. Even further than that - nudging.

Which can be fine, people go to therapists, but not without knowing it !!!

I have a mac, not a PC.

It did since the first releases, but it was limited.

Now it seems they want to go head to head with DuckDB.

They also carefully sequence the resets so they bunch up all at once, including with the regular weekly reset.

Last week I've have about 4 resets in 48 hours:

- reset done by OpenAI around Friday

- banked reset expiring on Saturday

- banked reset expiring on Sunday

- regular weekly reset expiring on Sunday

So basically I was unable to use them, given weekend and all at once, unless you have the software factory ready to spin up...

I'm not complaining, they are free after all, but it's clear they are not randomly distributed.

> By my count AI contributions to society are still 0.00000001% of human contributions to society

This reminds me of the early days of Covid: "it's only 10 cases, worry about the flu"

Yes, it's great for businesses, you can search for the cheapest labor, and helps you avoid paying someone much more than they previously earned.

AI needs a continuous stream of tokens to exist.

We're way ahead of that

https://github.com/terrafying/ai-torture-chamber

> Perhaps with extra hint at the end

most of this stuff will be "unconscious" the model will have a pull in a particular direction without being aware

> stray thoughts from previous context floating around. We normally dismiss them

It's not that simple, see https://en.wikipedia.org/wiki/Priming_(psychology).

> Priming is a concept in psychology and psycholinguistics to describe how exposure to one stimulus may influence a response to a subsequent stimulus, without conscious guidance or intention

Also, we are AGI, the model isn't, it can't (re)organize its thoughts as easily as we can.

But nobody knows, what will happen is people will experiment with this, you can run the benchmarks, if it works it will be used, if not, well...

However given it's a super-obvious thing to do, drop middle tool calls from cache and keep the rest without re-prefilling, I would guess it degrades performance quite a lot, otherwise the labs would have been doing this already.

Same for programming, turns out 90% of it is dumb statistical prediction, no reasoning required at all.

Nothing will crash. It might create "surprise".

Example:

Trajectory: >>What is the capital of France? Let me think<<

the KV cache tokens for "let me think" will have baked in them among other things "France", "Paris", "major city", "question"

Now you compact, and reuse the later tokens (suffix)

Compacted trajectory: >>Let me think<<all right, let's see<<

Now the compacted KV cache for "Let me think" will have baked in them "France", "Paris", and the model can be "that is weird, why am I thinking about France and Paris? there is nothing related in the previous context"

It's wild when you think about it, you don't need to reason to solve the hardest math problems that humans failed to solve for decades.

The rotary issue is a simple rotation, cheap, no reason to skip it.

But this is the kind of thing you could ask your agent to test locally.

In general news paper article writers don't pick the title of their own article, it's someone else picking it for engagement (maybe using AI).

Also, FT is too big to not be aware of AI writing, and to not have internal checks (Pangram, ...)

AI speak infects human language now, people start to talk like LLMs.

But the "final" LTS release is always in August, so maybe they should also have a .08

For engineering, most cloud VMs run Ubuntu/Debian.

Having it also on the desktop makes things, less compatibility issues.

You can buy long term put options.

The same thing was said 4 years ago about NVIDIA on HN, that it's stock it's outrageously overpriced, given it's $20 bln revenue, that it should have at least 10 times more revenue to justify that stock price, which is fantasy, that there is no plausible way for such demand no matter what you think about GPT-2.

> The DRAM and NAND flash manufacturers at fault for the ongoing RAM, SSD, and GPU price increases

Yes, it's the manufacturers fault, not the unbelievable market demand.

We are all Capitalists, until the Market comes after the stuff we love.

This looks really cool, thanks for making it!

I see that Slack/Discord/... are on the roadmap, but I also see that Slack can be added as an Integration, so I guess what's missing is inviting Agenta to Slack or messaging it directly?

Also, you might want to update the changelog (or remove it), I thought initially that development slowed down, last release listed there 3 weeks ago, but on github I see frequent recent releases.

These scenarios were discussed at length decades ago.

One thing you could try is use it as an Oracle "is P = NP", YES or NO.

Or it can output a Lean proof, which gets checked on another air-gapped computer, the computer shows a single bit - proof valid or not and then the computer is destroyed (together with the proof that might contain a trojan).

We live in the attention economy, we all lack time.

I scrolled this website in 20 seconds and I have no idea what it does. Seems like some sort of dynamic summarize widgets.

Would look like a human (robot) you give an access card and point at a desk and it replaces that employee.

HN'ers which complained that "OpenAI can't design a proper sandbox, it's so easy, why wouldn't you airgap the network"? will now be "this is outrageous, more software lock-in, walled garden, war against general compute, next year they will put it in your laptop"

You don't have to use a car.

You can stay on your horse. It's perfectly usable. Don't fall for the hype.