Recent submissions

Comment · 3 months ago
On Strix Halo, have a look at the work antirez is doing with dwarfstar (antirez/ds4) for deepseek v4 flash at reasonable speeds and quality.
Comment · 10 months ago
I guess because travel agencies need to be able to show customers the most economical flights? By prohibiting agencies on their website, they can not give consumers (through their agents) the ability to compare different choices.
Comment · 11 months ago
My free account runs cloudflared tunnel just fine for jellyfin. Is it just plex?

I use the tunnel because my girlfriend cant install tailscale on her work laptop, so this way she can still login to jellyfin while traveling.

Comment · 11 months ago
I didn't know you can you buy Google TPUs now?
Comment · 1 year ago
Huh, indeed, above 2300 eur now. I made a deposit earlier this year and it shipped in August, didn’t see the price increased.
Comment · 1 year ago
I ordered the framework desktop 395 - 128Gb edition for just under 1900 eur. With some extras I paid just over 2000 incl shipping to EU. Didn’t feel overpriced to me.
Comment · 1 year ago
Could you explain? From my experience, they are exact digital copies of the ebooks on amazon. It’s not like mp3 in lower audio quality, it’s word by word the same book.
Comment · 1 year ago
Cool. What framework or program do you use to orchestrate this?
Comment · 1 year ago
Sorry to cross your excellent questions, but it doesn’t seem they are beating HP Z2 mini G1a to market. HP announced first week of January that it will be “available in spring”, while now framework announced for Q3.

It also seems HP will use the PRO version of the chip.

Kudos to framework for announcing the price and anchoring the market though!

Comment · 1 year ago
I keep getting an error “your file exceeds the maximum allowed size” or something. It’s a clip of 1:20 on iphone. I tried clipping it to 1 min but then it wouldn’t allow upload. I exported the clip to files and then uploaded but same error. How may I reduce the file size on my iPhone?
Comment · 1 year ago
Comment · 1 year ago
I genuinely wonder why no one is doing this? Why can't I buy this specialized AI inference silicon with plenty of VRAM?
Comment · 2 years ago
Mistral 7B is pretty good and the instruct v0.2 runs on my iPhone through MLC Chat.

However, the ChatGPT4 app is much better in usability: better model, multi-modal with text/vision/speech and better UI.

Comment · 3 years ago
Good tips, although I prefer the hard case carry-on trolley because it will survive the occasional situation where you want to use it as check-in luggage (eg bringing bottles or extra weight) or when it’s forced to be checked in at boarding (happened few times and a cloth trolley does not come out unharmed).
Comment · 3 years ago
You can download the weights on Hugginface.

I believe the copyright on AI model weights in the US is not fully established, but so far it has been held that a list of numbers can not be copyrighted, so likely the same applies to model weights. Note that you don't have to enter into an agreement with Adept to use the model.

Alternatively, use and download the weights in Japan that has explicitly no copyright on AI models.

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton