Hacker Newsnew | past | comments | ask | show | jobs | submit | sigmar's commentslogin

Really? Maybe my brain is broken but I saw lots of forced analogies and LLM habits. Like where it does an em dash and a short list:

"I was getting steak ads everywhere—Instagram, news sites, even in places where ads should not be allowed to have that much audacity." ... "everything simply—salt and pepper, no fancy rubs, no sauce safety net." ... "with twine—very farmer’s market cosplay, very “trust me, I’m artisanal.”"


Before everyone piles on this comment with "whoosh" and "it was sarcasm" and such — have you noticed that reacting to ironic, sarcastic comments as if they were meant literally is what real LVL 80 trolls do oftener and oftener? On internet, you can never know who is pulling leg...

ebay started attempting consignment more than ten years ago, but I think lately they only do it for luxury items. Which makes sense to me as a lot of people would just send in junk. https://pages.ebay.com/ebay-consignment/

Neat research. I find figure 11 interesting. The models behave so differently there.

imo the benchmark should be named Can_It_Pull_a_CharDet_Bench


Scrambling? Seems to me xAI built too much capacity (for what they can use in 2026). Does that mean OpenAI built the right amount? I don't see how this proves that just because we see one AI company willing to sell compute. We don't even know the terms/pricing.

> Scrambling?

Yes.

To quote:

> Anthropic CEO Dario Amodei said his company tried to plan for 10-fold growth. But revenue and usage increased 80-fold in the first quarter on an annualized basis, which he says explains why it’s been so hard to keep up with demand.

> “That is the reason we have had difficulties with compute,” Amodei said Wednesday at his company’s developer conference in San Francisco. Amodei added that the company is “working as quickly as possible to provide more” capacity and will “pass that compute on to you as soon as we can.

https://www.cnbc.com/2026/05/06/anthropic-ceo-dario-amodei-s...

I think "scrambling" is a fair characterization of the CEO saying "we have had difficulties with compute" and "working as quickly as possible to provide more"

They've also signed new compute deals with Google and AWS recently.


for anyone wanting a glossary to explain the naming scheme here:

E4B = 4B effective parameters (using per-layer embeddings)

E2B = 2B (like above)

it = instruction tuned (rlhf and all that jazz)

assistant = Multi-token drafters (the new 2x speed up)


> assistant

naming still hard I see


I wonder if they hadn't decided to call it a drafter when they named the files and were using assistant internally? google being google...

Yes they should have stick with the naming convention.

google/gemma-4-31B-it-ass


>try them directly on Google AI Edge Gallery for Android or iOS.

I'm not seeing any update to the app on my android phone... maybe later today?

>We’ve published an in-depth technical explainer

I was expected a pdf link, but this goes to a brief article on twitter/X. lol, okay...



Moves like this make me wonder- What chance is there that these models are nationalized in the near future? What will happen to the investors/economy in such a scenario?

It's not even hypothetical. Once these systems reach a certain level of capability, they WILL be nationalized ("We'll take it from here, boys").

Nationalization often happens when growth ends. The Pennsylvania Railroad was private as long as the profits were rolling in. But once growth ended (because of cars and planes and buses and ....) the company went bankrupt. Then we ended up with Amtrak because the country needs a train system.

once it gets nationalized, it will be plagued from red tape. The model will likely look like how china is controlling their AI. It's not nationalized, but they have a complete tight leash on it

So nationalized models === more openly available and downloadable models? Seems the argument you're trying to make says "less leash" rather than a stricter one.


Nice! What search did you use?

What specifically is the goal of the pre-release review? Just to patch government systems first? Seems like the government was banning internal use of anthropic's models 2 months ago and now wants exclusive access for some amount of time. Clown show...

Study design, data collection, analysis, and peer review take time. O1 came out a little over 1.5 years ago

At this point the study is already mostly irrelevant because the model in question has long been far surpassed by new models. It seems traditional publishing doesn't work for really fast moving fields.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: