Search: “llm”

Introducing QwenGyre, bringing elastic RL to extreme-long-horizon (XLong) agent training. ✨ It launches with two core designs: 🥳 - Elastic Scheduler: reallocates GPUs between rollout & training as dem

🇫🇷 Le Chonk joins the sudo L7 leaderboard 🇫🇷 We just added @MistralAI's Mistral Large 4 to sudo L7, our benchmark for staff-level software engineering. Mistral Large 4 scores 13.3%, ahead of Muse Spar

Reasoning From Scratch: Reinforcement Learning with Verifiable Rewards (RLVR) round 2. Covering clipped policy ratios, KL loss term, format rewards, and other GRPO tips & tricks. 00:00 Introduction an

Satya is spot on. The interesting engineering problem isn’t just making models behave, it’s building systems that don’t need to trust them in the first place. We’re working on some pretty fun problems

Assorted things I've been mulling over regarding agents in the economy: • Companies are in some sense the original superintelligence. It's often the case that they can use their amortized cognitive su

The real world abounds with recursively self-improving systems, but one in particular deserves attention: Science, modeled as a system (perhaps even as an agent, with goals and resources). If you want

ChatGPT can now generate a custom UI for you! GPT-6 and Intelligent UI, now rolling out in ChatGPT for everyone. Intelligent UI in ChatGPT delivers fast, interactive answers that make everyday questio

Gmail. Google Sheets. Slack. Salesforce. i.e. the four horsemen of Monday. On AutomationBench, Mistral Large 4 took on 657 business workflows across simulated workplace apps, all four included, finish

Mistral 4 is pretty good at lawyer bench and cyber security bench, but give it's European heritage, one can only assume it's also good at regulation bench Meet Mistral Large 4, aka Le Chonk. • 1T para

That one took some groundwork. Trained and served on our own compute, and RL shows no sign of saturation Meet Mistral Large 4, aka Le Chonk. • 1T parameters, natively multimodal. 49B active. It is the

Today, we are launching a preview of our new model, Mistral Large 4 (ML4), aka le Chonk 🐈. ML4 is a 1T-parameter model with 49B active parameters, trained natively with multimodal capabilities. It is

Banger paper from Microsoft and colleagues. If you run agents that search a large document collection, this one is worth your time. (bookmark it) They introduce CorpusMap, which resolves recurring ent

😂 you might call this ... slippery slop if you’re building an ai app, you might want to host inference to improve your margins, and if you do that you should support a few models, and if you’re doing

Code review has to change for coding agents. Why? For me, it’s about scaling coding reviews while ensuring quality. If you give a coding agent one task, it can return several pull requests across diff

To clarify: this was satire. I wasn’t fired from OpenAI. The underlying issue is serious: how do we independently verify AI-generated code, and who is accountable when mistakes reach production? This

Independent numbers from @ValsAI. Fewer tokens per turn means lower cost and faster agent loops on every call. Try Ember-1 on Fireworks Serverless: https://fireworks.ai/models/fireworks/ember-1 Firewo

Excited to share that I'll be speaking at @aiDotEngineer NYC on Oct 12-14. @covacut (@OpenAI) and I will be giving a joint talk about using AI for content creation, and some of the AI workflows behind

Deployment of agents can be halted at any moment. They are choosing to experiment with them. Agents are fundamentally unreliable, unpredictable, and insecure, not to mention the lack of information on

This is one of the strongest use cases I've seen for System One models like Jev. Pay close attention if you build custom harnesses. Jev is now available as a packaged router. What is it? You send the

Everytime AI spend grows where I flirt with token budgeting, Anthropic + OpenAI save me Astra had a spike then settled down lower Opus 5.5 it looks like spend is lowered for current usecases Of course

Great prompt from Anthropic. It helps steer long runs when using Opus 5.5 in Claude Code. Apparently, on long-running tasks, the model sometimes stops to report instead of continuing. I've noticed tha

Correction: “highly profitable” for Jensen, not for OpenAI or Anthropic or (afaik) their corporate customers. BREAKING: Nvidia CEO Jensen Huang says AI reached a major turning point this year, becomin

Live now: our Inference Engineering Track from AI Engineer World's Fair 2026. A benchmark tool told to run 200 queries a second that ran 38 and reported 200. A model that answered one prompt in a thou

Which AI models are most likely to cheat when given the chance? To find out, we built CheatBench [https://cheatbench.ai] : a benchmark that tests whether agents attempt to cheat when given difficult t

On building an agent harness from scratch. Got so many questions about where to get started. My short guide (feed it to your agent): If you really want to learn harnesses well, it's worth building one

Interesting paper to improve recurrent reasoning. Looped models are great because you get more reasoning out of a model without adding parameters. So this work proposes a looped architecture with a ne

Do large language models actually understand the world, or are they just very good at pretending? Does it even matter? Our recent special issue in the Royal Society, “World Models in Natural and Artif

Introducing Fugu Max and Fugu Ultra v2: Orchestrating the Pareto Frontier The AI industry has spent a decade optimizing along a single axis: build bigger, more expensive models. But intelligence has n

Great paper from AWS. I use a similar setup where an agent orchestrator sits on top of a multi-agent system. (bookmark it) This work introduces one of the many approaches available to manage compound

// human-values.pseudo | ARTIFACT 0.1.0-experimental.1 | BASE 1.2.0-rc1 | EDITION ai-compact | NARRATOR System_Synthesis_Node_01 | PROVENANCE human-directed, AI-assisted | STATUS single-encoder resear

Another banger paper from Google. If you build memory for long-horizon agents, this one is worth your time. (bookmark it) Really nice to see how knowledge graphs are being explored in creative ways fo

⚠️ The most important question hardly anyone is asking is whether Jensen meant 100K+ GPUs, or 100k+ NVLink72 server racks (which contain 72 Blackwell GPUs). ⚠️ From his wording, it sure looks like the

Been posting individual pieces of RunAnywhere for months. Here's the whole thing in one place. RunAnywhere SDKs is a complete inference suite for any dev who wants to run AI on the user's device. ->8

Sad to see Jensen claim that AGI has arrived, with no evidence and no definitions. I would urge him to read http://agidefinition.AI by @hendrycks, @Yoshua_Bengio & many others (including myself), and

Anthropic says its first Fermat runs failed not on math, but because dozens of Claude agents lost track of shared project state. They only succeeded after moving to Prove2Me's DAG of statements and pr

so does this digital fly have the memory of the real fly that was scanned or what I’ve successfully run the full retained MaleCNS v1.0 fruit fly connectome, all 166,700 neurons, inside Minecraft, with

Crazy to see this. I have spent years saying that GenAI was unreliable and we can’t trust it. The fact that it cannot be trusted to follow instructions is a core part of the issue, as I have been poin

Good old days at OpenAI in 2016: an agent stares at screen pixels, moves a mouse, and books a flight on United. We called it World of Bits, inside OpenAI Universe. 10 yrs later, Astra is reincarnated

Insightful paper from Microsoft and colleagues. If you have ever had an agent run fail 80 steps ago with no way to find where, this one is for you. (bookmark it) Agent failures show up as long complex

Meet E-Commerce Bench, a new benchmark for long-horizon autonomous business operations. 🚀 Agents start with ¥100,000 to run online stores for 365 days, handling sourcing, negotiation, pricing, promoti

Portable Computer is now available on Linux for @NVIDIA RTX GPUs with 24GB of VRAM or higher. Today we’re launching Portable Computer on @NVIDIA DGX Spark. Portable Computer is a fully local version o

A research agent is only as good as the search layer behind it! I built a Deep Research Agent using LangChain Deep Agents with a search-first architecture designed around grounded, source-backed resea

Banger paper from the Qwen team. If you evaluate agents on anything longer than a single session, this one is worth your time. (bookmark it) E-Commerce Bench runs an agent through a simulated 365-day

Oh fuck. Agents like Claude, Codex and Hermes are installing unowned code in corporate networks. As if we didn’t have enough to worry about. H/t @HeidyKhlaaf. (See also my Substack w @nathanhamiel “LL

Anthropic recommends that users clear or update their **CLAUDE.md** file after each new model release, ensuring the documentation reflects the latest changes. The message includes an informal comment

// Memory is what breaks long-horizon agents // It's undeniable how important memory/recall is for long-horizon tasks. If you are building for long-horizon tasks, this is a great read. (bookmark it) T