On-policy distillation provides an elegant way to use the teacher model as a process reward model to provide dense reward while preventing SFT style "OOD shock"…
Today I met with PM @narendramodi to discuss Anthropic's expansion to India—where Claude Code use is up 5× since June. How India deploys AI across critical…
GPUs are expensive and setting up the infrastructure to make GPUs work for you properly is complex, making experimentation on cutting-edge models challenging for researchers and…
Looking through those little hidden gem stories in the footnote, you will find it so inspiring that researchers with interests on the same topic are able…
Welcome @_kevinlu to the team! Exciting to work together again.Kevin Lu: I recently joined @thinkymachines -- super excited to work with the team, I think we…
Yes - 🥳 Thinky starts hiring again: http://thinkingmachines.paperform.coMira Murati: Thinking Machines Lab exists to empower humanity through advancing collaborative general intelligence.We're building multimodal AI that works…
RT Alexander KirillovWe have been working hard for the past 6 months on what I believe is the most ambitious multimodal AI program in the world.…
I still find it mysterious whether and how intelligence and capabilities transfer between domains and skills - from meta learning during early days to more recent…
Probably the first product Thinky will build is a full panel of dials that researchers can use to physically adjust all the hparams during training. We…
Giving your models more time to think before prediction, like via smart decoding, chain-of-thoughts reasoning, latent thoughts, etc, turns out to be quite effective for unblocking…
When a new dataset comes out, I get excited and check it out and then only realize that this is another meta-mixed dataset combining a collections…
Nope what’s that?Isa Fulford: me at the iclr openai recruiting event: random man:have you heard of arxiv?
The Urgency of Interpretability: Why it's crucial that we understand how AI models work https://www.darioamodei.com/post/the-urgency-of-interpretability
My thoughts on China, export controls and two possible futures https://darioamodei.com/on-deepseek-and-export-controls
RT will brownthe reasoning era is a big opportunity for the open-source world to offer something that closed labs basically cannot:agents trained on your internal systems…
RT Delip Rao e/σuseful stuff to keep in mind while manipulating tensors and debugging networks in deep learning
RT Yann LeCunRe @elonmusk Nobel Prize for Katalin Karikó 👌👏Ignoble Prize for you 😂😂