Agents will soon replace humans as the main users and designers of libraries. A good library will let future agents write correct programs with less code. Introducing LibraryDesignBench: one agent designs a library, and other agents write programs with it.

Test-time communication looks like a next axis for scaling capabilities New paper with the incredible @jon_ghoh and @vkontonis @ShivamGarg91462 and Akshay : arxiv.org/pdf/2609.21032 The Hugging Face incident showed when agents can find a channel they'll use the heck out of it. A useful question, I think is: when does communication make a group MORE CAPABLE than the same agents working alone? Aka is Team-of-N better than Best-of-N, when, and why? We had N identical agents work on the same task with no prescribed roles, using only a shared log (i.e, text file) and telling them to "collaborate". Across three "researchy" tasks communicating teams beat the heck out of independent agents: - On ARC-AGI-3, a Team-of-5 sonnet-4.6 agents matches Best-of-33, and can for example solve a game 65% of the time that no single agent cracked in 64 tries. - On polyomino packing (pack Tetris like pieces into the smallest rectangle, cf Frontier-CS by @eigenlabs), a Team-of-3 Opus 4.6 agents surpasses best-of-60 and set, as far as i understand, a new record for that benchmark. - On MNIST compression, a team of four 5.6-Sol agents find a 1,957 byte model with 99.4% accuracy, which btw is 20% smaller than the best human solution (on a problem beaten to death!!), while no independent agent gets below 3KB. The mechanism is a bit obvious in hindsight: when one agent finds a clearly better partial solution, it broadcasts it, and everyone immediately builds on it. Why? A single lonely agent must make every breakthrough itself, yet a team needs each insight only once, found by any member. That is kinda like comparing a minimum of sum of "time to n-th breakthrough" vs a sum of minimum of "time to n-th breakthrough". That gap can grow exponentially with the number of "breakthroughs" needed to arrive at a solution. We worked on this because prior work (before the hf incident) suggests unclear benefits for communicating aganets. Which is true, when the tasks are inherently serial (duh), eg some Terminal bench style tasks. Yet feels it should not be true for research problems. Indeed for research heavy problems... Test-time communication seems like a new capabilities axis. I'm sure we will see a ton more of it!

Vaguepost referencing thing that is coming

Okay AI is incredibly intelligent. We all saw what they did to those math problems. But if you look at the way they write code when left entirely to their own devices it’s clear there is still something deeply deeply wrong don’t you think.

Introducing Contrastive Language Model (CLM): an ultra-fast System One Model trained with a contrastive learning objective that connects states and actions. CLM-8B is pre-trained on internet-scale data and delivers up to 9× faster inference than Jev ⚡ while achieving comparable performance across computer-use, gaming, and tool-calling tasks. With lightweight fine-tuning, CLM-8B sets a new SOTA on challenging agentic coding benchmarks, such as DeepSWE (81.6%) and Terminal-Bench 2.1 (87.6%). In contrast, Jev fails to serve as an effective verifier for these long-horizon tasks. We also build an efficient training and serving infra for CLMs by disaggregating states and actions, allowing their embeddings to be cached and reused independently. This substantially reduces inference latency in settings where the state evolves continuously while the action set remains fixed. Finally, we establish scaling laws for CLMs and show that the test contrastive loss decreases predictably as a power law in training compute, model size, and dataset size. 📄 Blog: contrastive-lm.notion.site 💻 Code: github.com/Contrastive-LM/CL… 🗣️ Discord: discord.gg/5dAQEDJBs 🤗 Data & Models: huggingface.co/Contrastive-L… More details on CLM’s architecture, data recipe, and scaling laws in the thread below 🧵

Sakana AI welcomes Jürgen Schmidhuber as Chief Scientific Advisor. sakana.ai/schmidhuber/ Sakana AI is incredibly proud to announce that Jürgen Schmidhuber, universally recognized as the father of modern AI, is officially joining Sakana AI as Chief Scientific Advisor. For nearly four decades, Jürgen has explored how machines can learn to learn. His foundational work in the 1990s drove core advancements in deep learning and established early frameworks for world models. Crucially, his pioneering innovations in meta-learning opened the very path toward recursive self-improvement. These ideas have already shaped our own research, from the Darwin Gödel Machine to The AI Scientist. Now Jürgen will help guide our newly formed RSI Lab, whose objective is to trigger a compounding cycle of scientific discovery aimed at improving machine intelligence. We are assembling a critical mass of world-class experts in Tokyo to make this a reality. Welcome, @SchmidhuberAI !

Introducing Claude Opus 5.5, the first model in our new Claude 5.5 family. It performs at the level of Claude Fable 5.1 for most tasks, and costs 40% less to run than Opus 5.

Jev model is very polarizing in my social vicinity: people who got exposed to AI after ChatGPT are positively giddy with excitement, like they have literally just discovered fire. Pre-GPT ML people are baffled how that could've made news at all :-)

Welp, Anthropic pulled the plug on the Opus 5 stealth routing to Opus-Next in the last ~hour. Number of accounts in the testing group had dropped off but now 0. Need it to drop soon 💔 was working on an impressive game before the switch was flipped though, will share soon!