I run 20 coding agents in parallel as my everyday workflow. Today, I’m launching Jcode. It’s an open-source agent 20x more memory-efficient than Claude Code, so you can run 20x more agents at once. Try it today:

My summer project is done! A 20 video, free course on post-training to accompany my book is all on YouTube with slides open for modification & re-use. ~12 hours of content covers the core foundations and some research areas I think will grow in importance. It was a fun time to review all the fundamentals again, as it is clear in the next 1-3 people the amount of people wanting to learn post training will likely 100X again from today, as we have already 100X'ed from two years ago. As AI agents get increasingly capable at coding and discussing these fundamentals (see the code exercises accompanying the book that I am refining with the community) I think developing clear intuitions for how models work and why is one of the most important skills going forward in AI. Still, learning the post-training math is the best way to battle test them. I personally just in this course am starting to master how forward/reverse KL relates to post-training topics. Thanks to all my viewers, and I'm happy to answer questions in the book discord or understand how to better teach the various reward models, on-policy distillation, new RL algorithms, etc. Plus, the book is 50% off right now with the code PBLambert on Manning to celebrate the launch. I'll share the relevant links below. Who's going to make this course for pretraining?

We re-tested GPT-5.6 Luna from @OpenAI on ARC-AGI (Verified) following its recent 80% price reduction: - ARC-AGI-2: 59.6%, $0.18/task - ARC-AGI-1: 90.7%, $0.07/task The new results match Luna's original performance at a much lower cost.

How does a model understand and explore a world it has never seen? We introduce 🔭VISTA🔭, a visual harness that gives a VLM long-horizon vision for reasoning in an interactive world. With Claude Opus 5.0, it reaches 100% RHAE on @arcprize's ARC-AGI-3, perfectly solving all 25 public games. Blog post: 🧵

What's the next OpenClaw moment? I asked @GregKamradt this when we chatted last week. "They're not spending money on their behalf. They're spending money on behalf of humans" "Agents aren't talking to each other that much yet. It's very single-player mode. My agents aren't talking to your agents. I would imagine they probably have something productive to talk about, but they're not doing it right now." "The persistent always-on agents are not here yet. It's still very prompt, go do a work task."

> Last year, I was constrained by tokens. I fixed that by joining OpenAI > Then I was constrained by CPU, now I feel my constraint is actually *attention* .@steipete came to agents/pizza/wine to demo his agentic engineering workflow

ARC-AGI-3 is built different, it has dumbfounded almost all regular attempts so far because it's so much harder than anything that came before. It has no rules, it's agentic and has no explicit goals, they need to be discovered. @tufalabs won the first milestone of @arcprize > There is no language built into the benchmark, but these guys "put the language back in", because in their view - it's the best way to climb up the notional "abstraction mountain" and effectively use many of the abstractions which have evolved over millions of years of language evolution. > They built a novel harness "The Duck" around a 27B open weights model (Qwen 3.6) to solve extremely challenging and novel reasoning problems that require abstraction. > This is the launch video of their winning agentic harness, "The Duck". We have also released an exclusive interview with them on MLST, just dropped. > The million dollar question is: what will @fchollet think about how they've done it, and is this a step towards AGI?