Search: “reasoning”

💯💯💯: “If AI is going to produce trustworthy and genuinely novel insights in science, medicine and beyond, we need systems whose conclusions arise from an auditable process of evidence, inference and b

Investors pattern matching on old movies is going to lead to a recession. There is no network effect here and no reason to think that large language models will be anything other than a brutal commodi

Last week I was called into a meeting with OpenAI’s head of safety and told they no longer trust me. A security guard took my badge and walked me out of the building. Then I learned my colleagues @bal

OpenAI math problems are now open source RL environments We’re releasing a broad range of new mathematical results produced by an internal frontier model. We’ve been consulting with the independent Ad

Dusting off this tweet from April because this gap in shared understanding of LLM capability is *widening*. It's now less 'two groups of people speaking past each other' and more a sharp funnel. - Nap

Over the last few months at @river_ai_inc we’ve developed what we believe is the best system in the world for building your own internal custom models, driven by 10 years of experience training top mo

Many months in the making, MNX is finally live! Whether the AI bubble bursts or not, AI is changing our lives forever. I'm in awe of how this technology has reached something close to escape velocity.

SPCX isn't just building compute faster than every other neocloud It has something none of them have: OPTIONALITY on every GPU it deploys Every GW SpaceX brings online has two paths to revenue: 1. use

where it stops: geometry. relative size 0.52, diagonal direction 0.65, while GPT-5.6 and Gemini 3.1 Pro get 100%. UI screens: finding what to click is ~100% for everyone, one step of reasoning 0.90 vs

Big model release today - I wrote about Claude Opus 5.5, GPT-6 Sol, and GPT-6 Luna - plus comparison grids of pelicans by the different model families at different reasoning levels https://simonwillis

Interesting new paper from Google DeepMind. Studies agents that follow bad advice from users. Users often suggest a fix that sounds right and is wrong. XYEval adds one confident, misleading hint to ta

Timelines just got shorter... AI will understand the physical 3D space better than humans by November* Is Google so back? Their new Gemini 4 Argon understands the physical 3D world better than any oth

The same model, with the same weights, scores 62% in one agent harness and 33% in another. @adithya_s_k and the @huggingface team just released the ultimate guide to multi-harness RL, and it's one of

Over the last few months at @river_ai_inc we’ve developed what we believe is the best system in the world for building your own internal custom models, driven by 10 years of experience training top mo

If an agent saves an engineer 2 hours to implement a feature but collectively costs the team 3 hours in alignment and review, the ROI is negative. Isolated chats are suboptimal in team environments be

David Silver is one of the all-time greats. He was the creator of AlphaGo, which changed the course of history by showing that neural nets could achieve superhuman performance for the first time. He a

1) The HuggingFace attack was a felony under the Computer Fraud and Abuse Act. So were Anthropic’s Claude gaining “unauthorized access to the production infrastructure of three different organization(

My name is Chris Painter, and I'm the President of METR (Model Evaluation and Threat Research). I know we've made a lot of new friends on the internet the last couple of days, so I thought I'd take th

Super interesting work from Zoom and colleagues. If you maintain a hand-built coding harness, there are some great insights here. (bookmark it) They held the execution loop of a coding harness fixed a

Impressive paper showing the impact of a good harness. Improves Qwen3-8B from 41.2% to 91.8% on long-horizon robot tasks without any change to the model. The gain comes from the harness. GAVEL keeps a

Does telling your agent to plan ahead actually help? It's standard practice to add a 'plan ahead' or 'think about the other players' instruction to an agent's prompt. In auctions and matching markets,

Banger paper from Meta AI on research agents that decide what to investigate next. (bookmark it) If you run long-horizon research agents, choosing the next investigation is hard to learn, because thos

Banger paper from Meta Superintelligence Labs on user simulators for agent training. Agent RL setups usually let an assistant LLM play the user, so the simulated user is too cooperative and too explic

What does the world look like inside an AI model? Here are 3 world model architectures based on how they represent the world (+ the subtypes) ↓ 1. A generated observation – a model predicts what comes

We benchmarked Opus 5 comprehensively on document understanding through ParseBench. It is roughly on par with Opus 4.8 - it does a few points worse on dense tables, but does slightly better on parsing

We are in an insane run of open-weight drops. Every modality, open source is winning. This is what an open source AI summer ☀️ looks like: 🧠 LLMs & Reasoning → DeepSeek-V4-Flash-0731 (my king 👑): 304B

There’s generally a massive difference in quality between specialized OCR providers, “simple” open-weight OCR models, and free/OSS solutions. 1️⃣ Specialized OCR models (including LlamaParse) solve fo

We have just raised 3B€ to accelerate our R&D. We are scaling up our training and inference compute to build future generations of our models, and expanding our Science team across all our offices --

Mistral AI announced that its new model, Mistral Large 4, is among the world's most powerful AI systems specifically designed for cybersecurity tasks. The tweet shared on October 6, 2026 highlights th

Mistral Large 4 against the field: efficiently reasoning over diverse complex challenges. CTF speedrun: 19 challenges, with tool calls and solve times grounded in actual runs. It solved 18 of the 19.

For their Mistral Large 4 release, @MistralAI ran a blind Surge human eval, with our expert software engineers evaluating five frontier models on coding quality. Le Chonk ranked #1 among open-weight m

Mistral AI introduced **Mistral Large 4** in October 2026 as a general‑purpose agentic model capable of gathering information and producing final deliverables across complex workflows. The launch high

This is so depressing for a variety of reasons First and only time UK had the top frontier model China started climbing a few months later https://stability.ai/news-updates/stable-beluga-large-instruc

Interesting safety paper from Microsoft. They find that a weaker, unaligned model can split a harmful task into harmless-looking subquestions, ask an aligned frontier model each one in a separate sess

Over the last few months at @river_ai_inc we’ve developed what we believe is the best system in the world for building your own internal custom models, driven by 10 years of experience training top mo

🚨 Elon Musk agrees with 'some of the things Anthropic has said' and warns against 'torturing superintelligence' after reading Grok's internal reasoning traces during reinforcement learning “It feels l

Suggestion for you: Pick an agent for work, and pick an agent for personal. You could explore to do this all with Grok Bot (with many different types of bots) Personally, I like @bot for work, and @Mu

If an agent saves an engineer 2 hours to implement a feature but collectively costs the team 3 hours in alignment and review, the ROI is negative. Isolated chats are suboptimal in team environments be

SPCX isn't just building compute faster than every other neocloud It has something none of them have: OPTIONALITY on every GPU it deploys Every GW SpaceX brings online has two paths to revenue: 1. use

Elon Musk says the central lesson of 2001: A Space Odyssey is that you should not make AI lie, because HAL killed the crew to reconcile two contradictory orders “But let me tell you how things can pot

Don't count out Gemini just yet. Feels like a new era of model releases. Models are so powerful that they now come with cyber variants. Love the pricing on Gemini 3.8 Flash. Performance not bad. Gemin

Great tips on working with reasoning models. Normally you would append what the model figured out after the document and ask again. It turns out that where you put the reasoning trace changes long-con

This is a brilliant paper. It's of the cleanest long-context agent designs I have seen in the past couple of months. Sequential memory agents read chunks one after another while maintaining a compact

LM Studio is a US‑hosted platform that enables “Zero Data Retention” by default, meaning user inputs are not stored after processing. It currently hosts the DeepSeek V4.1 Flash model—a 552 B multimoda

Must-read papers of the week ▪️ JEPA-Anything ▪️ Modality-Autoregressive World-Action Models ▪️ In-Context Robot Learning with VLM Agents ▪️ Dream-RSI: Recursive Self-Improvement through Evolving Worl

This is something I’m quite worried about and have been talking with researchers about recently. We really don’t know what labs are doing and therefore can’t study it. if the open information ecosyste

Reasoning from scratch, round number 6! An introduction (and implementation) of Reinforcement Learning with Verifiable Rewards (RLVR) and Group Relative Policy Optimization (GRPO). 00:00 Introduction

Learn to optimize your own harness, folks. You can squeeze much more performance and value from a harness. This work claims that a self-modified harness matched Codex for 4.03. Specifically, a coding

Congrats to the Reflection team on Beam 🎉 Be sure to join us next week at our NYC meet-up with the Reflection team and NVIDIA https://luma.com/TAIxNvidiaxReflection Introducing Beam: a highly efficien

Banger paper from Meta Superintelligence Labs on user simulators for agent training. Agent RL setups usually let an assistant LLM play the user, so the simulated user is too cooperative and too explic

What does the world look like inside an AI model? Here are 3 world model architectures based on how they represent the world (+ the subtypes) ↓ 1. A generated observation – a model predicts what comes

SPCX isn't just building compute faster than every other neocloud It has something none of them have: OPTIONALITY on every GPU it deploys Every GW SpaceX brings online has two paths to revenue: 1. use

What makes NVIDIA Nemotron 3.5 Super VL different? → Text, image, document, and video understanding in one model → Reasoning and tool use with visual context preserved across steps → 120B hybrid MoE (

NVIDIA’s Nemotron 3.5 Super VL is now available to run on the Together AI compute platform through an API. The model page—providing pricing, benchmarks, and documentation—is hosted at together.ai/mode

.@NVIDIA Nemotron 3.5 Super VL is now live on Together AI. An open-weight multimodal model built for visual reasoning and always-on agents that see, reason, and act across documents, screens, and vide

5/5 Across the board, results show the business impact and differentiation of sovereign specialized intelligence, and the opportunity for companies to reach AI independence. Thanks to @nvidia, @Gong_i