
After a long wait, releasing the SYNTH paper! It’s not pretraining, mid-training or post-training, it’s just training: a fully synthetic single-stage pipeline to train workable reasoning models with unprecedented data efficiency.

building open ai infrastructure @pleiasfr — χαλεπὰ τὰ καλά

After a long wait, releasing the SYNTH paper! It’s not pretraining, mid-training or post-training, it’s just training: a fully synthetic single-stage pipeline to train workable reasoning models with unprecedented data efficiency.

*we knew
if you need a gpu we know a guy

It’s actually funny how models are still trained in a world where they had no widespread effects. Like Claude is very happy to correct typos unprompted, unaware that with the deluge of spotless slop it’s no longer signal.

if you cannot be first on raw capability, actual product is user alignment (or processing scale, but this one is for homemade stuff)

well, tipping point where glm has replaced openai as my secondary coding agent.

still stuck in the data mines, but i guess nothing ever happens.

just make it believe it’s in the sandbox
Claude can't see Reddit, b/c Reddit blocks all LLMs except Gemini, b/c they have a Google contract. I ask Gemini for posts matching criteria, but Google blocks direct quotes. So I ask Claude to use a headless browser, which it refuses for 'safety'. Fuck everything about this.

By Austegard on the blue site muninn.austegard.com/blog/ed… Currently too deep in synth data mines to give it a try.

Even Monad got brainwashed now.
We asked GPT-6 Astra to edit Qwen3.5-0.8B weights to improve math. It directly changed only 0.016% of weights, fixing 0.8B's reasoning loops & improving math up to 6x and even instruction following 3x, without gradients or training. We call this Weight-Space Programming: frontier LLMs can now natively read/write neural network weights! 🧵 Blog: rikiyat.github.io/blog/weigh…

so i guess, quite a few people rediscovering eternal truths about hardware boom/bust cycles


Also reminder K3 is a custom license that requires a dedicated agreement with full disclosure past 20M revenue (though after testing it a bit, unlikely to be much of a concern).


Users cannot "reverse engineer the Model or attempt in any way to derive the Model source code, parameters or architecture" and "benchmark the Model for publication without Giotto.ai’s prior written consent". Wonder why.
Today we announce the launch of the Giotto family of models. We are unveiling Giotto Alto and Giotto Solo, the smartest European sovereign AI models on the market. Giotto Alto is our flagship model built for advanced reasoning capabilities, free for everyone and available on the App Store and Google Play. It is hosted on European cloud infrastructure, so your data stays in Europe. Giotto Solo is the smartest model that runs on a single GPU, so organisations can keep their AI on their own infrastructure. We are a Swiss lab with a European heart and we built this so that Europe can run advanced AI on its own terms. Use Giotto: cloud.giotto.ai/

anthropic definitely needs to buy more sheet music environments (either that or the taskers lack taste)

some teasing from london
Since SYNTH is travelling across Europe this week: coming to London on Thursday for the first session of AI research Foundry, in excellent company (w/ pretraining/data team from Gemini and MAI-1).