johanlinden.bsky.social - Profile | ThreadSky | a Reddit-style client for Bluesky

Krea AI has released its first image model: Krea 1 Ditch heavy prompts: generate photoreal portraits, origami cats, or surreal parks in one click. All tools in-box: 4K upscaling, style transfer, 1000+ presets.

submitted 5 days ago • 1 comment

Just a reminder of the ridiculous talent of this guy.

submitted 55 days ago • 0 comments

"On Tyranny" is a #1 NYT bestseller again. I wish the moment were different. But I’m glad the book is useful. And grateful for all the kind words about putting the 20 lessons to work. snyder.substack.com/p/twenty-les...

submitted 64 days ago • 465 comments

submitted 64 days ago • 1 comment

ahahaha brilliant

submitted 66 days ago • 15 comments

New from 404 Media: we've seen an ICE database that filters people by hundreds of different, highly specific categories. Experts say possibly helping ICE identify and deport people. Scars, location data, immigration status, eye color, much more. Created by Palantir www.404media.co/inside-a-pow...

submitted 67 days ago • 14 comments

Booker finally yields after more than 25 hours of speaking 👏👏👏

submitted 75 days ago • 618 comments

20 Lessons on Tyranny by Timothy Snyder Read by John Lithgow "Now, more than ever, we need the wisdom of our intellects, the patriotism of our citizens, and the passion and talents of those who still believe in the American experiment." Share widely. youtu.be/cXR5HLodsT8?...

submitted 76 days ago • 6 comments

NEW: Family of Mahmoud Khalil just released footage of his arrest “Stop resisting” “He’s not resisting” Agents seen continuing to ignore Mahmoud’s wife as she asks basic questions

submitted 93 days ago • 704 comments

In this interview with Stephanie Hare for BBC, I shared my recent thoughts on the progression of models in programming and reasoning, and especially on how the next frontier, agency, will force us to quickly develop technical and societal safeguards. www.youtube.com/watch?v=3cFk...

submitted 97 days ago • 1 comment

submitted 97 days ago • 771 comments

This is Marco Rubio explaining how the USA promised to defend Ukraine forever if they got rid of their nuclear arsenal left after the Soviet Union fell. This is why lil marco was sinking into the couch. He was hoping we wouldn’t find it…so don’t RT right now this very second.

submitted 105 days ago • 2059 comments

Welcome European Research Council (ERC) @erc.europa.eu to Bluesky 🦋

submitted 128 days ago • 0 comments

Two weeks ago, reasoning models (strongest at hard problems) were only available for subscription Now, if you don’t want to pay, you can get reasoners for free from Microsoft Copilot (o1), ChatGPT (o3-mini), or DeepSeek (r1). The best reasoners (o1-pro & o3-mini-high) still require a subscription.

submitted 135 days ago • 9 comments

Released a new Python package today. Hope it helps some of you out! github.com/mikeckennedy... #python

submitted 135 days ago • 5 comments

How far behind are Chinese LLMs? When looking at training compute—a key metric of AI model development—we found that Chinese language models have scaled more slowly than their global counterparts since late 2021.

submitted 144 days ago • 2 comments

Remember when we were arguing about the term "foundation models"? It feels like ages ago! With those base models culminating in DeepSeek v3 in Dec 2024, 2025 will likely be the year of LLM specialization! Anyway, I will be writing about it more soon :).

submitted 139 days ago • 23 comments

Deleted a post on Janus 7B because it is unfair to compare a diffusion model to a multimodal model. Its not a great image generator compared to small diffusion models👇, but until we find out parameters from something like Aurora, it is hard to compare

submitted 139 days ago • 0 comments

gentle reminder that deepseek is an open-weight model and NOT open source. there's nothing open about it; no information on how its trained or data trained on, just like its western counterparts

submitted 140 days ago • 13 comments

Yes, DeepSeek R1's release is impressive. But the real story is what happened in just 7 days after: Original release: 8 models, 540K downloads. Just the beginning... The community turned those open-weight models into +550 NEW models on @huggingface. Total downloads? 2.5M—nearly 5X the originals.

submitted 139 days ago • 3 comments

The Philosophy of Data with @sabinaleonelli.bsky.social, @vlastasikimic.bsky.social, and Rhaul Sam #PhilSci #PhilSky

submitted 161 days ago • 2 comments

Sigrid Hjertén (1885 – 1948) was an influential Swedish modernist painter #WomensArt

submitted 164 days ago • 0 comments

I wrote a thread about public service broadcasting in an era of social media billionaires. One thing I forgot: get rid of public broadcasting and small countries like Sweden can kiss goodbye decent local radio and TV news. No profits, so no private company would pay for that. Ever.

submitted 166 days ago • 1 comment

On the left, my image. On the right, the Al image that was clearly trained on, and designed to plagiarize, my image. All Al generated imagery is inherently theft. It's unethical, environmentally disastrous, and does irreparable harm to both artists and society.

submitted 174 days ago • 338 comments

The devastation on its way is unimaginable - like nothing humanity has ever witnessed before - and everyone will wish they had been climate activists and ‘hysterics’ when they had the chance. We are in a moment of unprecedented stupidity

submitted 175 days ago • 747 comments

🌟 Spoiler: no need to code! #OpenDataEditor is our new open source app that makes working with #data easier for people with little to no technical skills. Join us for the release webinar! 🗓️ December 19 🕐 13:00 UTC 👩🏽‍💻 Online ✍🏽 Register: https://buff.ly/3OZ6sGv

submitted 181 days ago • 0 comments