gcr

This is oversimplified. The proposed question is whether a model with fewer parameters could achieve performance on one language similar to that of a larger model that’s been trained more broadly, which isn’t straightforward to do.

This isn’t technically true. Most model providers don’t send the thinking tokens anymore, so if you switch from one provider to another, you will be missing large parts of the conversation.

Ed’s interview covered many different topics. As an aside, he was asked what he thought of Apple’s position. The juicy side quote is now what defines the interview.

That’s my whole point! It’s ported to unsafe rust, not default rust.

The whole point of the rust rewrite is that bun is now thought to rely on rust’s memory safety features, but that assumption doesn’t hold if everything’s inside an unsafe block.

Palantir pioneered forward-deployed engineering, aren’t they deeply embedded in the above?

It’s my understanding that bun was ported to unsafe rust, so even these gains would require additional effort on the team’s part, right?

There’s a small albeit nonzero chance that frontier labs want Simon to stop using the pelican thing so they can draw attention to better benchmarks

This is an amazingly tone-deaf response to what is clearly an personal artistic project with a lot of heart behind it.

I agree with your last sentence, but we can’t partition the budget up so cleanly though. Societal spending is interdependent and additive across domains. Spending on cheap access to safe water or food reduces healthcare costs. Spending heavily on education tends to strengthen government in the long run, and better-run governments amplify efforts to improve housing/food/water/.

The one exception is AI. Historically, redirecting charitable funds towards AI safety tends to starve or undo the rest of these efforts, which is why I’m so disappointed by many EA institutions dropping other initiatives to put their eggs in the AGI basket.

Wonder what would happen if a hacker focused all ten thousand of them on a single area for an hour or two. Sounds like a really energy-efficient way to demolish a city.

X Wing: Wedge’s Gamble (1996) by Michael Stackpole shows the rebel alliance using similar tricks during the battle of Coruscant.

But in the public image, the EA community is synonymous with doubling down on AI / AGI to the exclusion of the other projects.

OpenPhil changing its name to Coefficient Giving, 80000 hours and bluedot and (to a lesser extent) CFAR dropping other initiatives and switching to AGI promotion… to my knowledge GiveWell is the only other big name that continues to advance other initiatives. Then look at figureheads like SBF committing fraud and begging for a pardon from the architects of the USAID shutdown… We begin to paint a picture of a community that’s (by and large) abandoned its principles for power.

I know the view from the inside is more nuanced, but I think it’s a reasonable association for random members of the public to make.

My critique of the EA community is that it’s myopic and unregularized. If you really think AGI is make-or-break for civilization, it’s completely rational to deprioritize side bets.

Wouldn’t turning off swap fix this issue?

when it comes to video gaming I’ve found Bazzite to be generally far less fiddly than windows 11, surprisingly

The supported method to get a new one each boot is to truncate the file to 0 bytes and disable systemd-machine-id-commit.service

Double-check that this method actually works though.

Machine ID is used for things like dhcp leases, log rotation, etc. IPV6 addresses or transient MAC addresses are derived from it

there are very strong graphical glitches on iOS, it looks like every other fragment isn’t being rendered or something based on the camera angle

could the author clarify what we should be seeing or what the point of this is?

Reframing this point: Some good books aren’t borrowed because they’re not discoverable, not because they’re boring.

The library is highlighting a few titles for increased visibility to ask, “would this pique a reader’s interest if they knew about it, or is this generally bad?”

Without this stage, the library would expunge more genuinely interesting titles.

I’ve always kinda felt the role of a library is for recall rather than precision

Good points, makes me reconsider. Than you. 100% keep going—this is how you learn! I hope my complaining about the commoditization AMA individualism of AI doesn’t add too much elitism or discouragement to the discussion, though rereading my post I definitely see how it could.

In a sense, sharing products in threads like this helps future people see what else has been done!

It’s insidious! I used Claude code at my last job enough for it to influence my writing and speaking style without me really noticing, even though I tried to be wary of that.

Oh pardon, I’m trying to sarcastically complain how a lot of comments in this thread have a similar form: “I used this pattern in my own agent, which is different from (all the other agents which use the same representation)”

Agentic development tends to encourage siloed individualistic development, so a lot of engineers reinvent similar patterns from first principles. It’s easier to write your own new thing than survey other approaches, so you’re more likely to perceive good ideas as original to your session.

Agree with your critique. I think this work is presenting common ideas as novel without thinking through existing problems. Defining a provider-agnostic event graph that enables full session branching replay was the whole point of pi: https://mariozechner.at/posts/2025-11-30-pi-coding-agent/ , though the language around it perhaps didn’t click until a bit later. I don’t even think pi was the first to do this.

Another critique: the abstract mentions how their system allows for “branch[ing] a run at any event without re-executing the shared prefix,” but that’s only possible with very careful KV caching. Generally, rerunning inference from an earlier point still incurs O(n) input token cost and this paper is working at the wrong layer to see that. In this work, execution refers to tool calls but token generation is the expensive part.

shh you’re ruining the fun! :-)

Very cool work!! This is the same pattern we used at $MY_STARTUP to develop $MY_HARNESS which persists the entire graph to disk, unlike all the other agent harnesses which only store the graph nodes and edges.

Event graphs aren’t just the agentic foundation for $MY_HARNESS — they’re the working cognitive substrate, native to what our favorite toolcall gremlins actually consume.

(Looking for lead investors for our angel syndicate btw! DM me if interested)

Every desktop environment is a massive memory hog. Do you really want something minimal like xfce on a touchscreen?

Something’s way off with these numbers. The page says it encodes video at 640MB/s which is quite large even for 4D data and doesn’t match the filesize of the demo splat (7.4MB / 2sec, or ≈3.4MB/s).

In fact they say the raw file size of the demo splat was only 427MB, so maybe the 640MB/s was a statement about encode speed? Why write it that way instead of “this demo splat was encoded in 0.6sec” or even just “the time to produce the original splat took longer than the time to encode this video format”?

um, no? “The government” owning a stake in OpenAI is different from “the people” owning a stake.

How are you reading that differently?

posters upthread are talking about comprehension and value systems, not literacy.

"functionally illiterate" is the brush that one paints with when describing people of opposing political viewpoint or lower socioeconomic status, for example.

This page starts flickering madly when I pinch-to-zoom. Until a11y details like this are figured out, I don’t think this should be considered for general use beyond a cool prototype.

I could write a program to generate the first 100MB of pi in a couple kilobytes. That certainly counts as “data compression” but isn’t useful outside this particular problem instance.

Now you understand why I thought something written in good faith might be interpreted as shade. Happens all the time.