NEW Stanford & MIT paper on Model Harnesses. Changing the harness around a fixed LLM can produce a 6x performance gap on the same benchmark. What if we automated harness engineering itself? The work introduces Meta-Harness, an agentic system that searches over harness code by exposing the full history through a filesystem. The proposer reads source code, execution traces, and scores from all prior candidates, referencing over 20 past attempts per step. On text classification, it improves over SOTA context management by 7.7 points while using 4x fewer tokens. On agentic coding, it outperforms all hand-engineered baselines on TerminalBench-2, scoring 37.6% versus Claude Code's 27.5%. This is a big deal! Here is why: The harness around a model often matters as much as the model itself. Meta-Harness shows that giving an optimizer rich access to prior experience, not just compressed scores, unlocks automated engineering that beats human-designed scaffolding. Paper: https://t.co/hqkZaWbBTl Learn to build effective AI agents in our academy: https://t.co/1e8RZKs4uX

@gpt_alex I moved to Paris and picked up smoking and then I lived in Japan I spent a summer living on a farm in upstate New York I started spearfishing in Florida https://t.co/L6r2MpTTmE

If you know what these things are and want to go to Monterey Bay with me, hit me up https://t.co/tI99d2Drlt

Meta just released the Efficient Universal Perception Encoder on Hugging Face A vision backbone for edge devices that unifies image understanding, vision-language modeling, and dense prediction via multi-teacher distillation. https://t.co/qnF84e5t09

“Grok Analysis” is one of the most useful features on 𝕏. Tap the Grok icon on the top right of any post, and you’ll get an instant Grok analysis. https://t.co/PTyaprV0f8

Grok Imagine "extend from frame" 30 second continuous video.💫 https://t.co/W1WllENYGm

Here’s what Grok can do • Talk about anything: ask real questions, get honest answers, almost zero boring policy warnings • Search super fast: latest news, 𝕏 posts, or the whole web in seconds • Run code live: math, science, data, games, whatever… solved on the spot • Create pictures and videos: just describe an idea and it’ll make (or edit) image or clip • Handle real life: brainstorm businesses, study help, stock advice, savage roasts • Real-time & fun: generate memes, vulgarly roast someone (with their photo), bedtime stories, you name it Like a super smart friend who actually gets it Try it @Grok 🔥

Nvidia CEO Jensen “@Tesla stack is the most advanced autonomous vehicle stack in the world. I’m fairly certain they were already using end-to-end AI. Whether their AI did reasoning or not in somewhat secondary to that first part.” https://t.co/YJAlQJybgx

Everyone thought the future was carbon fiber Elon Musk looked at the physics and chose stainless steel for Starship instead Sounds insane.... until you realize stainless gets stronger at cryogenic temperatures, handles reentry heat better, and costs massively less than advanced composites. It doesn't even need paint He chose a material that is faster to build, easier to weld, tougher in extreme conditions, and built for rapid iteration Classic Elon: ignore convention, trust first-principles engineering, and pick the solution everyone else missed He is taking science fiction and making it real. Building things that only existed in imagination, and pushing them to the absolute limits of physics

Hi yeah I have a question @elonmusk what did you do to Grok Imagine??!! The Detail!! What is going on!! ❤️ Look at this!! Its amazing Geek mode unlocked Thank You!! https://t.co/nKKbn20xyp

Elon Musk: It’s remarkable how much people stick with their ideological tribe, no matter the evidence. “It is remarkable how much people believe things simply because it is the belief of their in-group, you know, whatever their sort of political or ideological tribe is. There's some pretty hilarious videos of some guy going around as a racist Nazi or whatever, and he was trying to show them the videos of the thing that they are talking about where he is in fact condemning the Nazis in the strongest possible terms and condemning racism in the strongest possible terms. And they literally don't even want to watch the videos. People, or at least some people, will stick to whatever their ideological views are, whatever that sort of political tribal views are, no matter what. The evidence could be staring them in the face and they're just going to be a flat earther. There is no evidence that you could show to a flat earther to convince them the world's round because everything is just a lie, the world is flat.” From: All-In Podcast November 12, 2025

@elonmusk Grok Imagine will also add English if you prompt for it. imagine prompt: In a scene reminiscent of Conan the Barbarian's epic fantasy world, a powerful and stunning woman exudes authority from atop a majestic throne. Her presence is both commanding and serene, evoking the untamed beauty and strength associated with Conan's universe. Flanking her on both the right and left are sleek black panthers, their muscular forms at rest yet brimming with potential energy. These majestic beasts serve not only as protectors but as symbols of her formidable power and connection to the wild. The air is charged with an ancient magic, as if the very scene is a gateway to realms untold, and she, its sovereign ruler, possesses the wisdom and ferocity of the ages. Ghost in the Shell 1995 cel animation style by Production I.G and Mamoru Oshii, authentic cel grain texture, classic 90s cel shading, extremely desaturated cold palette, dominant muted dark purple tones, only faint cyan-green accents, no vibrant colors, no bright neon, atmosphere oppressive melancholic existential, spectacular volumetric edge lighting in muted purple with soft rim light and subtle cyan highlights, very deep volumetric shadows, sharp precise linework in Oshii's direction, masterpiece quality, pure 90s OVA aesthetic, hyper detailed, cinematic composition, no chibi, no flat shading, no cartoonish gloss, no generic anime exaggeration. speaking japanese:「ネットに潜れば、真実が見える。でも見えすぎると、もう戻れない。私はとっくに戻る場所をなくした。」   Add subtitles in english.

Grok 4.20 Beta ranks #2 with 97% accuracy score on the 𝜏²-Bench for Telecom (Agentic Tool Use) It outperforms Claude Opus 4.6(max), GPT-5.4(xhigh), and Gemini 3.1 Pro, while closing in on GLM-5 scoring the top in agentic work flow Tool calling is the whole game for AI agents, and this is where Grok 4.20 takes over with state-of-the-art intelligence that fires up instantly, making it the fastest at tokens per sec in the industry

Updated the product growth stacker app we now have ability to create variations of the same roadmap. I know April is around the corner and chances are, your roadmaps have things pushed/pulled or you didn't even start yet. Signup https://t.co/OOj3MPpyhn #buildinpublic https://t.co/1i4yv1t5V1

KwaiKAT presents KAT-Coder-V2 Presents KAT-Coder-V2, an agentic coding model that performs close to Claude Opus 4.6 https://t.co/8d346gFZY7

proj: https://t.co/Kj2iaq1DNH abs: https://t.co/3E5MxHROAq

LongCat-Next: Lexicalizing Modalities as Discrete Tokens - Matches or beats SOTA across multimodal benchmarks - SotA audio: strong on both recognition and TTS accuracy - No trade-offs: adds vision/audio without hurting core language performance https://t.co/g2LKPI6mnp

abs: https://t.co/kTjlXqjAeu repo: https://t.co/nKBA33NxYh hf: https://t.co/hwTtl4pPRk

+1 more

PRBench: End-to-end Paper Reproduction in Physics Research - Presents a benchmark of 30 expert-curated tasks spanning 11 subfields of physics - All agents exhibit a zero end-to-end callback success rate https://t.co/kJf4ToAWjt

daVinci-LLM: Towards the Science of Pretraining - Matches larger model perf with half the size - Huge reasoning gains: +23 pts on MATH, strong code + science scores - Quality > scale: smarter data (not more data) drives major performance boosts https://t.co/6IMp7ABa3M

abs: https://t.co/Er7GOpxwG1 repo: https://t.co/BEgftxMzC9 model: https://t.co/cW6redVodO data: https://t.co/xVhWUeBlGZ

+2 more

Bass Windu is becoming an AI masterpiece. I want a full movie. https://t.co/uVzRk5q030

@emollick Yes, pubmed started automatically pulling from the major preprint servers in 2023: https://t.co/9Mc6rwQKQ9

Warren Buffett: "Most of the time, the Fed is not that important." "The Fed is of enormous importance during a panic. People tend to hang on their every word in between [panics], but we don't pay any attention to it." https://t.co/xpDEwUmjjp

Tau Bench got an update! Tau Bench is one of the most adopted Agentic Benchmarks. They now added “Banking” a fintech-inspired customer support domain built around a realistic knowledge base of 698 documents across 21 product categories. Tasks require agents to search this corpus, reason over what they find, and execute multi-step tool calls. "There's this transaction I want to dispute. I also want to file a credit limit increase request." The best model achieve 25% success of tasks and ~< 10% on pass^4

Claude vs. Claude Code vs. Cowork. If you've been confused about which one to use and when, this post will clear that up in under two minutes. Anthropic now offers three distinct ways to interact with Claude, and each one targets a fundamentally different workflow. Think of it as: Chat for thinking, Code for building, and Cowork for doing. Here's a quick breakdown: 1️⃣ Claude Chat This is the conversational AI assistant most people already know. You type a prompt, Claude responds, and you iterate together. - Turn rough ideas into structured plans through conversation - Write emails, reports, essays, and long-form content - Research and summarize complex topics in minutes - Analyze documents, PDFs, and images - Build interactive prototypes through Artifacts The key here is that everything happens through conversation. You're thinking with Claude, not delegating work to it. It's available on every device, has a free tier, and supports persistent memory across sessions. The tradeoff is that it has no direct access to your local files (upload only), and it can't generate raster images natively. 2️⃣ Claude Code This is a terminal-native coding agent. You describe what you want in plain English, and Claude reads your codebase, writes code, runs tests, fixes errors, and ships the result. - Build and debug entire features across the full codebase - Write, run, and fix tests automatically - Manage git workflows and create pull requests - Spawn multiple parallel agents working on different parts of a task simultaneously It handles the full development cycle end to end, from planning to execution to testing. With the CLAUDE(.)md configuration file, you can teach it your project's conventions, patterns, and constraints so it writes code the way your team expects. The tradeoff is a steeper learning curve compared to Chat, and token costs can add up during heavy sessions. 3️⃣ Claude Cowork This is the newest addition. Anthropic describes it as Claude Code for the rest of your work. It's an agentic desktop assistant that automates file management and repetitive tasks through a GUI. You describe an outcome, and Claude plans, executes, and delivers finished work: formatted documents, organized file systems, spreadsheets with working formulas, and synthesized research. - Direct local file access and editing (no upload/download cycle) - Schedule recurring tasks automatically - Assign tasks remotely via Dispatch from your phone - Computer Use lets Claude control your screen directly It runs inside a sandboxed virtual machine on your computer, so Claude can only access folders you explicitly grant. You don't need to know how to code to use it. The tradeoff is that your computer must stay awake for tasks to run, and it's still in research preview. Here's how to think about choosing between them: → If you need to think through a problem or get writing/research help, use Chat → If you're building software and want an autonomous coding partner, use Code → If you have a clearly defined deliverable that involves local files and desktop workflows, use Cowork All three are included in the same subscription starting at $20/month, which makes it one of the highest-leverage subscriptions in productivity software right now. I've put together a visual below that maps the workflow of each product side by side. If you want to go deeper into Claude Code specifically, I wrote a detailed article covering the anatomy of the .claude/ folder, a complete guide to CLAUDE(.)md, custom commands, skills, agents, and permissions, and how to set them all up properly. Link in the next tweet.

Charlie Munger's last Interview Before His Passing at 99; https://t.co/Oym1zPBBLy

Charlie Munger's last Interview Before His Passing at 99; https://t.co/Oym1zPBBLy

If I could give my younger self one piece of advice, it would be this: Your capacity to learn is far greater than your current knowledge. Many people let the need to be "smart" get in the way of open-mindedness. But if you can view life as an adventure and approach disagreement with curiosity instead of anger, you'll find yourself evolving to higher and higher levels.

Marc Andreessen says AI is the "silver bullet excuse" for companies laying people off, but most layoffs are actually due to higher interest rates and overstaffing during COVID: "This entire labor displacement thing is 100% incorrect. It's completely wrong. It's classic zero-sum economics." "It was the combination of the two—interest rates going to zero during COVID, and then the complete loss of discipline at all these companies when they went virtual and when employees just became an icon on a screen." "What you have happening right now is that essentially every large company is overstaffed. We could debate how much—it's at least overstaffed by 25%. I think most large companies are overstaffed by 50%. A lot of them are overstaffed by 75%." "And now they all have the silver bullet excuse—it's AI." @pmarca with @HarryStebbings