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

I think the funding OpenAI can obtain will decrease.

Claude shipped a lot of books from Taiwan to the United States.

It feels like Claude will be crashing more and more frequently from now on.

Error code 502 --- what the admin panel

We were doing some infrastructure maintenance. It's back now!

I don't understand. What difficulties would these Chinese companies face if they held their training overseas?

The US could impose secondary sanctions to close the loophole, or pressure these countries behind closed doors to not allow these data center uses. Since the only goal of the export restrictions is to hamper Chinese AI efforts, that would be a logical step. If there is one thing that is certain then it's that the current US administration doesn't care at all about open trade and international trade agreements.

The input can be text, images, audio, video, and PDF. You can also set the resolution (low/medium/high, and ultra_high for images) to control costs, which is great.

After my research, I found that if you want to use this skill, it's best to specify the range of browsers you want to support in the AGENTS.md / CLAUDE.md of the repository (e.g., "Only use the Baseline Widely Available feature"), and keep the experimental features disabled by default. This way you can avoid some worrying issues.

What I need most right now is a model like TypeSafe's Jev that can be built directly into the computer.

Jev allows for numerous small judgments: Is this email important? Which program should handle this request? Is this operation risky? There's no need for large LLMs to slowly generate long paragraphs of text; all you need is a fast, structured result that returns the message.

Having a local Jev-like model in the operating system for quick decision-making, and then calling the larger model when deeper thinking is needed, would be very useful.


I'm using Ollama https://ollama.com/library/kev for this now.

Not recommended to use Ollama, I like Unsloth.

https://sleepingrobots.com/dreams/stop-using-ollama/


> There's no need for large LLMs

Do you know how big jev is? Decision models are small but the "smart" (aka useable) ones aren't that small.


I think ~30b params

They are even small enough to run in browsers: https://openjev.com/

https://github.com/TheoLeeCJ/SemIf-OpenJev

dabbling with it in browsers but i expect it would answer the mail for you, if you find it accurate enough


In a way you want something like blackberry messenger, a centralised hub for all app messages, notices etc, and then JEV hooked into it as a traffic controller of sorts. I wonder if there’s anything like that, perhaps on Linux?

In Taiwan, this was news from a year ago. But why Jensen would come forward to admit it now is what people find strange.

This is a fantastic website. Currently, it primarily displays satellite positions, but I think it would be even better if it could be displayed or filtered by orbit type. The current results clearly show the differences between different orbit types—the Central Park observations you provided already clearly demonstrate these differences—the Sentinel-2 satellite passes by around 10:50 local time (sun-synchronous orbit), the GOES-19 satellite continuously observes from geostationary orbit, and the International Space Station/Tiangong satellite passes by at different times each day (inclined low Earth orbit). This website would be even more educational if it could color-code or label the satellites according to orbit type (sun-synchronous orbit/geostationary orbit/inclined low Earth orbit/medium Earth orbit), or even label the local time for nodes for sun-synchronous orbit satellites.

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

Search: