MABU

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…

⚡Iterating at lightning speed! Gemini 4 Carbon is SOTA across key benchmarks and combines frontier capabilities with deep safety & security research.

Announcing three major scientific discoveries, one each in mathematics (harmonic analysis), computer science (complexity theory), and physics (quantum field theory). AI starting to help us improve our understanding of the universe! openai.com/index/building-sc…

p still probably doesn't equal np

There is no more code for us to write. The robots have won.

silicon valley literally had everything

Real life, baby

Today we’re releasing Muse Spark Open, the most capable US open-weight model to date. Built for coding and agentic work, it excels at long-horizon tasks, complex software engineering, and multi-step reasoning. Key Model Specs: → 816B total / 42B active parameters → 1M token context window → Natively multimodal across text, image, video, audio

🎯 Gemini 4 Argon has a 15% hallucination rate on AA-Omniscience, compared with 51% for GPT-6 Astra. 📊 It ranks first on the Vals Index, reflecting its high performance in legal and other knowledge work. 🔐 For now, only cybersecurity defenders in Google's Fairwind program can use it. Full story in our weekly AI newsletter, The Batch 📬 hubs.la/Q04zX8YW0 #DeepLearningAI #AIEngineering #Cybersecurity

New on our Engineering blog: Creating in-house source control. In May, we merged around 2,500 code changes per day. In September, we merged more than 30,000 per day. As a result, we became bottlenecked by developer infrastructure built for human (rather than agent) scale. Here, we share how we built a new version control system—and fully migrated our monorepo from git—in five months: anthropic.com/engineering/re…

You can now train your own Decision model like Jev locally! We increased Qwen3.5 0.8B’s aggregate accuracy from 20.7% to 74.3% across 3 decision benchmarks - on just 4GB VRAM. Turn any LLM like Qwen3.8, Gemma 4 into decision models with our open-source Unsloth repo. We fine-tuned with a Clef head using Unsloth and LoRA (r=64) for one epoch, increasing downstream accuracy from 30–37% to 78%. GitHub: github.com/unslothai/unsloth Guide and Notebooks: unsloth.ai/docs/basics/train…

We’re releasing a broad range of new mathematical results produced by an internal frontier model. We’ve been consulting with the independent Advisory Group on Mathematics and Artificial Intelligence at the Institute for Advanced Study, and we have drawn on their advice and public recommendations to inform how we release these results. github.com/openai/math

Introducing Griffin, the first model to pass the video Turing test. 48% of people who talked to it live thought it was a real human. Previous systems have had a pass rate <3%. It is #1 on NVIDIA's benchmark for full-duplex AI video. It’s the first Human Interaction Model (HIM).

Introducing Gemini 4 Argon – our new frontier model. It’s built for complex workflows across coding, enterprise knowledge work, and cybersecurity defense – rolling out today to a set of trusted testers through our Fairwind Program.

Introducing Claude Sonnet 5.5, the second model in the Claude 5.5 family. It’s a clear upgrade over Sonnet 5, runs more than 30% faster, and costs up to 30% less for most work.

It's a skill issue. Most graduates are not employable. I spoke about The “Classroom Student” vs “Out of Classroom Student” Theory of Learning at the Africa Data Science conference earlier this year. Good grades are not enough. I interview many of these grads and it's really bad.