JacobAsmuth

7 ms·

1.
by JacobAsmuth 12d ago
Except there's 10 trillion gears
2.
by JacobAsmuth 13d ago
How can I be more specific with no information?
3.
by JacobAsmuth 13d ago
Sure I do. You can just apply for access. What's your use case?
4.
by JacobAsmuth 13d ago
LOL
5.
by JacobAsmuth 13d ago
Astra is clearly able to acquire new knowledge in context and apply it. It was the whole thing that his ARC-AGI benchmarks have been measuring. It's a direct refutation of the original comment.
6.
by JacobAsmuth 13d ago
All those stupid Bell Labs researchers not inventing Uber or Tinder. How come they didn't just build the obviously popular and profitable businesses that became possible once they invented the internet?
7.
by JacobAsmuth 13d ago
It could also mean that Google can absorb essentially unlimited demand spikes by load shedding.
8.
by JacobAsmuth 14d ago
You're telling me for only 5x the cost and 1/10th the speed I can use a Chinese model which performs worse than Gemini 3.8 Cyber? And I get to do all the hosting and setup work myself instead of just using a model and framework wh
9.
by JacobAsmuth 14d ago
I'm confused. This doesn't mean they trained on it.
10.
by JacobAsmuth 15d ago
Will someone flag this comment please.
11.
by JacobAsmuth 21d ago
They'll be able to buy them without paying NVidia's 80% profit margin
12.
by JacobAsmuth 23d ago
The Googlers must be vague posting about something internal.
13.
by JacobAsmuth 23d ago
"No evidence" -> Large lab saying that they use distillation when training their models
14.
by JacobAsmuth 23d ago
https://techcrunch.com/2026/04/30/elon-musk-testifies-that-x... Make sure to stay updated!
15.
by JacobAsmuth 25d ago
Unlikely. Much more likely is that Opus 5 was trained in an RL environment with subagents, and it learned to talk this way when reporting progress to the invoking agent.
16.
by JacobAsmuth 25d ago
Not a risk.
17.
by JacobAsmuth 27d ago
https://www.yahoo.com/news/articles/ai-boss-trump-hates-beca...
18.
by JacobAsmuth 1mo ago
It has double the active params.
19.
by JacobAsmuth 1mo ago
You're suggesting that if a very good and cheap AI model came out tomorrow everyone would rush out to rent Azure instances to run batch size 1 inference on their model?
20.
by JacobAsmuth 1mo ago
"eventually" is actually a function of frontier model capabilities. You only get Qwen6-27B when you have Opus 7 producing extremely high quality tokens for them to train on. So the market for local models is always significantly b
21.
by JacobAsmuth 1mo ago
Ed Zitron has not yet made a single correct prediction about AI :)
22.
by JacobAsmuth 1mo ago
Yeah but Apple pays Klarna to provide that service. Circular financing!
23.
by JacobAsmuth 1mo ago
Wait until you hear about how airlines work, you'll start babbling about the rewards-points bubble
24.
by JacobAsmuth 1mo ago
> The research team, which includes collaborators from NVIDIA and Flower Labs, have come up with a new method for recursive self-improving AI agents to continue improving themselves. What happens if you apply the method to non-recursive
25.
by JacobAsmuth 1mo ago
Weird then that they don't get it all correct.
26.
by JacobAsmuth 1mo ago
Literally nothing indicates we've hit a peak, but I guess I'm discussing this with someone who thinks every iteration since Opus 4.6 had zero ROI so there's probably not much common ground here.
27.
by JacobAsmuth 1mo ago
Strong words coming from a blob of oxygen, carbon, and nitrogen.
28.
by JacobAsmuth 1mo ago
Well sure but you're running on tens of thousands of dollars of hardware.
29.
by JacobAsmuth 1mo ago
Per token. You need to look at pricing per task.
30.
by JacobAsmuth 1mo ago
Isn't the top end of the distribution by definition what it's capable of?

More ›