conception

It's a US group that made it, in the same vein that OpenAI and Anthropic have US models.

Depends on what you're using it for. DSf is pretty good at executing planned coding tasks but it isn't very intelligent. K3.1 doesn't need to be a coding master if it can be a fable replacement.

Can they fix that it runs like trash on 16gb of ram? That’d be cool.

I suppose think about it like you’re the captain of a ship. Imagine a captain always complaining about their peers, the ship, their outfit etc. And when giving orders they sound out their whole thinking process about why they are doing x or y. Versus never complaining or explaining in public. Certainly the former would be a less inspiring and weaker leader.

They aren’t attacking regulation but full time employment. They spent a tremendous amount of money in CA to pass a law that’s nearly unable to be repealed to avoid having their gig folks count as employees. That’s the attack - everyone being a 1099. Not regulated industries.

If this worked then a team of five fable models should crush al of its benchmarks, no?

Depends on where it is as many of their cameras are on private property.

And 10x the time.

I think agy has a “questions pending “ pattern for waiting questions rather than popups. I don’t like it but sounds like they were solving your issue

Omg this is amazing and solves so many of the issues I have with Yoto!

The top 0.001% has about 6% of the planet's wealth. ~60K people have a tremendous influence on politics, media and political discourse. Just a staggering amount of resources and power. Hardly an "out-group".

Whataboutism like this always misses the scale of it. Imagine Breaking Bad but with Crypto - suddenly all of Walt's cash logistics problems go away because doing crime in cash is not the same as crypto - the scale and ease are very different.

If you leave agents to figure it out, they will spend time figuring it out. If you give agents run-books and roles, they will align with the role at once.

That’s because the ultra wealthy are funding new talking points.

I’m sure carbon is great but also Google stock just took a pretty big hit so…

I love this never-ending people having vastly different experiences situation with LLMs. In my experience, Gemini flash 3.8 is an exceptional model and is the bulk of my agentic use as someone who bounces around models as soon as they come out. I would say it’s the best flash model until 1017 came out. It might still be better than 1017.

Gemini 3.8 Flash is so good in Pi. It may not show up in benchmarks but consistently good performance.

Baseten has flash at 800t/s.

You are missing a critical third - people organize and via unions or the polls push back and require less work. There is no reason we aren’t working seven days a week beyond people fighting not to.

I'll note - it's possible to get something out of LLMs that's readable, e.g. ADRs4AI is a method plus tools for recording the decisions you make while working with an AI coding assistant. As you work, the AI writes decision records into your repository. When it needs your input, it writes a question with options, a recommendation, its reasoning, and its confidence. You answer, often in one word, from VS Code or a phone app. The reasoning and your answer stay in the repo, so future sessions, with you or another AI, can see why things were done.

But it doesn't come out without the context to guide it (prompting/skills/etc).

People are trying to explain but it doesn’t seem to be clicking. I’ll give it a try and why LLM slop is so hard to read -

Your decisions start compounding. This keeps the why, so it compounds.

These are two sentences from the post that are supposed to be heavy hitter marketing lines but they are completely unable to be parsed in English. This is the gotcha of LLMs - it looks like words but there’s no actual writing ability here - just next token prediction.

Decisions don’t compound. You might say the the outcomes of poor decisions compound or something like that but these sorts of sentences are the uncanny valley of writing - it seems like writing but something is very wrong with it and that’s because it’s not actually stating anything.

Likewise “This keeps the why, so it compounds “ again - not actually a thought. What is compounding? What is growing here? Also no one talks like this, mainly because this isn’t actually conveying information.

The reason people hate AI slop is because at first glance it seems like you’re reading something that has something to say but you quickly find it’s an incredible amount of work to parse it due to poor readability, sentence structure etc and then further discover- oh there was nothing to gain from that effort in the first place. So the habit becomes “Ah slop? Not worth the time. Close tab.”

Our pediatrician told me something that always stuck - “You just need 1% more willpower than they have.” Always made setting boundaries a little easier.

I feel like also all the code optimizations make it treat all language like code and that produces patterns that align with how it generates code. Earlier models that focused on language seem to have done better - 4o seems like language wise it’s pretty good but advancements elsewhere make it difficult to use.

Yeah I have a perfectly fine iPad air and pro that are completely unusable. Nothing installs and nothing supports whats installed.

Well there we go. TIL.

That seems pretty arrogant that a hundred million years or more won’t restart life back on Earth. But I also might just be ignorant. What could we do to end all life on earth and prevent future life from forming?

Based on trying to buy anything retail that is scarce, bot detection systems are definitely not working. I wonder how their system would do against https://refractbot.com/ or other popular retail bots.

https://en.wikipedia.org/wiki/Fairness_doctrine Worked pretty well

If you run a readability analysis on slop it scores very poorly. LLMs generally continue to be mediocre with the veneer of quality without context to guide them.

On some models you can control the temperature so it responds the same every time to a specific prompt but the “randomness” of llms is usually a feature not a bug.