I have tested the ternary model on English/Italian and I'm surprised by how good the model is, also it's very fast.

Don't editorialize the title of the articles.

These are post-training reinforcement learning steps.

Yes, updated the submission title to say "post-training" to hopefully prevent further confusion

I don't think a bunch of similar tasks can really saturate the "create a SVG of X", because the model should have a quite good spatial understanding of the world and how everything interacts.

For example Gemini 3.8 Flash seems very impressive at first glance but the results are not actually very coherent, this shows that its "world model" is not particularly great (compare to SOTA models).

Gemini 3.8 Flash results are often not very coherent but it does put a lot of shading and details to hide the fact.

Just tried on Astra low and it gave 4.

I tried too. It got it. Maybe more importantly, who cares?

For example, I'm a nerd. I'm bad at baseball. I lack that kind of intelligence, even though it's more common than the ability to program. That doesn't also imply that you can't trust my Python code.

I used the 6 Pro setting in the chat. Too smart for its own good, I guess.

If you have never seen the game before probably.

Gemini 3.5 pro is never going to be released, it was a failure.

I think that would be more normal than comparing an LLM policy to having lost a child.

One is just something more people say, it's still as much unhinged.

Firefox really struggle with demo pages of text-to-video models because of the large numbers of videos in the page in my experience, this page seems to work quite fine for me tho.