Technology

This magazine is from a federated server and may be incomplete. Browse more on the original instance.

GolfNovemberUniform , in New AI image generator is 8 times faster than OpenAI's best tool — and can run on cheap computers

This is what happens when you let the community improve and modify your program

Fiivemacs , in Honda built a powered chair to zoom around theme parks while wearing an AR headset.

I will never move over to let them go by.

No_Eponym OP ,
@No_Eponym@lemmy.ca avatar
Fiivemacs ,

So I should move?

No_Eponym OP ,
@No_Eponym@lemmy.ca avatar
STOMPYI , in Honda built a powered chair to zoom around theme parks while wearing an AR headset.

People should learn how to close their eyes and find peace. All these distractions hitting up our ANS fight flight response is leaving its mark.

QBertReynolds , in 1-bit LLM performs similarly to full-precision Transformer LLMs with the same model size and training tokens but is much more efficient in terms of latency, memory, throughput, and energy consumption.

Says 1-bit then goes on to describe inputs as -1, 0, or 1. That's 2-bit. Am I missing something here?

will_a113 ,

It’s actually 1.58bits weirdly. The addition of 0 here was the significant change/improvement in this experiment. The paper isn’t too dense and has some decent tables that explain things fairly accessibly.

kevlar21 , in 1-bit LLM performs similarly to full-precision Transformer LLMs with the same model size and training tokens but is much more efficient in terms of latency, memory, throughput, and energy consumption.

Why use lot bit when one bit do trick?

tubbadu ,

Bits together weak

Fades , in India’s plan to let 1998 digital trade deal expire may worsen chip shortage

Fuck India, always on the wrong side

henfredemars , in NVIDIA GeForce Now gets pre-roll ads for free users

I mean, it is a free product. This doesn't bother me too much. Eventually they have to make money to stay in business as much as I despise Nvidia.

It would be a completely different story if I was paying for it.

geekwithsoul ,
@geekwithsoul@lemm.ee avatar

I agree that as a free product, the ads are less loathsome. But from a company that just posted a 265% increase in revenue, it’s not like they’re worried about keeping the lights on https://www.cnbc.com/2024/02/21/nvidia-nvda-earnings-report-q4-2024.html

tabular , (edited )
@tabular@lemmy.world avatar

Nvidia is making plenty of money. This is about having power of their user's computing.

onlinepersona , (edited )

A free product you pay for when you buy your graphics card?

Edit: oooh, I completely mistook the products. nvm

CC BY-NC-SA 4.0

Heavybell , in NVIDIA GeForce Now gets pre-roll ads for free users
@Heavybell@lemmy.world avatar

Huh, didn't expect geforce now to have a free tier.

WolfLink ,

It’s always had a free tier. The free tier existed before the paid tier did.

autotldr Bot , in India’s plan to let 1998 digital trade deal expire may worsen chip shortage

This is the best summary I could come up with:


India's plan to let a moratorium on imposing customs duties on cross-border digital e-commerce transactions expire may end up hurting India's more ambitious plans to become a global chip leader in the next five years, Reuters reported.

In place since 1998, the moratorium has been renewed every two years since—but India has grown concerned that it's losing significant revenues from not imposing taxes as demand rises for its digital goods, like movies, e-books, or games.

Reuters reviewed the letter, reporting that the WSC warned Modi that ending the moratorium "would mean tariffs on digital e-commerce and an innumerable number of transfers of chip design data across countries, raising costs and worsening chip shortages."

Studies suggest that India should be offering tax incentives, not potentially threatening to impose duties on chip design data.

ITIF's goal was to evaluate "India’s existing semiconductor ecosystem and policy frameworks" and offer "recommendations to facilitate longer-term strategic development of complementary semiconductor ecosystems in the US and India," a press release said, partly in order to "deepen commercial ties" between the countries.

The Prime Minister's Office (PMO) has also reported a similar goal to deepen commercial ties with the European Union.


The original article contains 482 words, the summary contains 194 words. Saved 60%. I'm a bot and I'm open source!

catch22 , in Reddit is licensing its content to Google to help train its AI models
@catch22@programming.dev avatar

Wheres my cut?

grayman ,

You're the product, not a customer.

jaybone , in China invents most powerful detonation engine for hypersonic flight

Oh it’s Lemmy.ml

TexMexBazooka ,

Yup

autotldr Bot , in Stability announces Stable Diffusion 3, a next-gen AI image generator

This is the best summary I could come up with:


Stability says that its Stable Diffusion 3 family of models (which takes text descriptions called "prompts" and turns them into matching images) range in size from 800 million to 8 billion parameters.

The size range accommodates allowing different versions of the model to run locally on a variety of devices—from smartphones to servers.

Stability has made a name for itself as providing a more open alternative to proprietary image-synthesis models like OpenAI's DALL-E 3, though not without controversy due to the use of copyrighted training data, bias, and the potential for abuse.

We do not have access to Stable Diffusion 3 (SD3), but from samples we found posted on Stability's website and associated social media accounts, the generations appear roughly comparable to other state-of-the-art image-synthesis models at the moment, including the aforementioned DALL-E 3, Adobe Firefly, Imagine with Meta AI, Midjourney, and Google Imagen.

While Stable Diffusion 3 isn't widely available, Stability says that once testing is complete, its weights will be free to download and run locally.

"This preview phase, as with previous models," Stability writes, "is crucial for gathering insights to improve its performance and safety ahead of an open release."


The original article contains 614 words, the summary contains 193 words. Saved 69%. I'm a bot and I'm open source!

huginn , in China invents most powerful detonation engine for hypersonic flight

This article is bullshit: they're reporting that a blueprint was made. No demonstration, no proof of it working.

Rotary detonation isn't a new concept and GE has demonstrated a rotary detonation scramjet for 0 - Mach 5.

https://www.geaerospace.com/press-release/other-news-information/ge-aerospace-demonstrates-hypersonic-dual-mode-ramjet-rotating

If you really think that 0 to Mach 16 is within reach with current technology RDEs I have a bridge to sell you.

Darpa has all the money of God and they've been throwing billions at the exact problem of rotary detonation. Everybody knows that detonation is more fuel efficient than deflagration and that turbine engines are holding jet aircraft back from hypersonic flight.

When they make a Mach 16 demonstrator I'll believe them. Until then I'm still very excited for the demonstrated and actual capabilities of RDE Scramjets.

pastermil , in China invents most powerful detonation engine for hypersonic flight

That's nice...

...when I see it in production.

Norgur , in China invents most powerful detonation engine for hypersonic flight

Yeah, no. Just no. Absolutely no. Faster than everything else.and higher than everything else and all of that by burning less energy? Come on people, stop falling for this kind of crap.

electricprism , (edited )
  • All
  • Subscribed
  • Moderated
  • Favorites
  • random
  • technology@lemmy.ml
  • updates
  • programming
  • tech
  • testing
  • marketreserach
  • drbboard
  • til
  • wanderlust
  • bitcoincash
  • Sacramento
  • All magazines