Search: “reasoning”

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 --

NEAR AI Lean agent solved ALL problems in Putnam Bench at 1/250 the cost of the second cheapest submission. $111 for 672 problems from the hardest undergraduate math competition in the world. Fully op

An effect of the rapid acceleration of AI is we are losing an empirical handle on what is happening in the actual micro-processes of work in the post-2026 long-running agentic era. We have some academ

Despite everything that has happened, OpenAI has seemingly decided to stay competitive by burning down the only meaningful bit of safety assurance we actually have today - CoT monitoring. Completely d

Y’all gonna get mad if I said I told you so … Researchers argues OpenAI and Anthropic will never get us to AGI. 21 top researchers from Stanford, Oxford, DeepMind, CMU & Meta released a paper saying L

Banger paper from Google DeepMind and colleagues. (bookmark it) A model reads its entire KV cache on every generated token, even though it ends up attending to a tiny slice of it. In other words, if y

cool new responses api features releasing today with astra: 1. async function calling — model can keep going while your tool runs 2. mid-turn steering — you can inject messages into the model while it

Just because someone puts the word AGI in their benchmark doesn't make it even remotely a reflection of what the general public understands by this term. As a layman, I define AGI as a virtual charact

Of all the use cases for GPT-6 Astra, I'm most excited for scientific discovery. We at @OpenAI have not pushed it to its limits on math and science. I look forward to waking up every morning and seein

A lot of hype around OpenAI's Astra model here on my timeline today. Apparently, this goes back to a new article from The Information, which said Astra is a 'recurrent depth or looped transformer'. It

I want to prevent a race into unmonitorability kicked off by confused reporting. The depth of the computation graph for our present frontier models, including Astra, is within a factor of two of GPT-4

Super interesting paper from Meta. Long-horizon research agents are coming. But one common problem with research agents today is the lack of originality and how to decide what experiments are worth ex

That doesn’t mean you can’t get value out of Kimi or Grok or whatever, but it involves you actually knowing some stuff about how to optimize your setup, and a willingness to change direction to switch

Very special Training Data drops today: Making Cities Awesome ☺️ We haven’t featured an application company on Training Data in a while. The diffusion of AI is the Most Important Thing, and the highes

you can now build @LangChain agents w/ agentic video understanding! google is consistently at the forefront of multimodal processing; this is the video analog of 'agentic vision' token usage + cost is

Great paper from Google and colleagues. It proposes an interesting approach to improve agents on long-horizon tasks. Long-running agents slow down and start poisoning their own context. Both symptoms

Interesting paper from Tencent. Tencent trains an agent to manage its own working context, and assigns credit at the level of individual context edits. Long-horizon tasks force a model to retrieve, in

As I've been saying for months, we are truly not ready for persistent agents. 3 comments I want to make about this article: 1) We don't have the full picture It's a great summary of what happened in t

LLMs will always be bulls in a China shop. Nailed it, December 2022. We are still suffering the consequences. And they are about to get so much worse. No wonder Silicon Valley tried for so long to dep

Obviously Mythos 2 already took over the editor desk at all the news organisations Overall, I’m very surprised at how little media coverage there’s been around the OpenAI / Hugging Face attack. It’s c

GLM-5.3-Flash is live on Fireworks on day… 2 Why? Because we take quality very seriously. We found a benchmark discrepancy we couldn’t explain, so we delayed the launch to investigate. Day 0 (Wed): we

Agentic coding lets developers—even those who previously focused on front‑end or mobile work—delegate code generation to AI agents and assume broader full‑stack roles. However, understanding core soft

API is live on QwenCloud: https://www.qwencloud.com/models/qwen3.8-flash 🙌Let's build something! ⚡Meet Qwen3.8-Flash, a multimodal MoE and an early preview of the Qwen4 architecture, now open-weight!

Impressive work from Alibaba. (bookmark it) If you build long-running agents and keep rewriting your memory schema, take a look at this approach. It basically treats agent context management as a prog

// Applying Anthropic Primitives at Large Enterprises // Frontier models collapsed the cost of writing custom code. But they did not reduce the cost of reviewing and maintaining it. Each bespoke solut

This is one of the most useful writeups I have seen on keeping an LLM judge effective in production. (bookmark it) Netflix runs judges over hundreds of thousands of show-level recommendation explanati

Banger paper from Google DeepMind. It's on the very hot topic of model routing. Routers assume the value estimate is free. Working out which specialist handles a query best costs money too, and it is

There are two separate questions regarding Dario Amodei’s post about AI and biology: motivation and reality. Motivation. Dario started in biology (was a PhD student of the great Bill Bialek). It might

1/2 Thanks Gavin for an especially thoughtful exchange. I don't usually spend much time on social media but I wanted to engage here because it really brings out the heart of an important conversation.

// Actions Speak Louder Than Words // Multilingual agent evaluation compares final answers and throws the trajectory away. The trajectory fixes cost, latency, failure mode, and auditability. New resea

Four small architecture decisions can cost up to 47% of a model's long-context performance. New research from Ai2, Carnegie Mellon, and the University of Washington isolates them. Normalization, GQA,

The future of FDE work seems closely related with all work around evals/posttraining/RL envs. FDEs are effectively responsible for the following: 1. Define the business problem. 2. Codify the business

Hey peeps. I know you're tired of low quants giving hard to believe numbers. I'm quite skeptical too and from what I tried I'm often left with the impression that the claims fall short. So this model

An accurate characterization of the arc of AI is that it is shaped by two trends: 1. Moving more and more logic to a neural model for tasks where training data can be densely sampled (e.g. the shift f

Fully onboard with productizing hillclimbing as an automated service for any agentic task. I've had the fortune of knowing @silennai since the AutoGPT days, and I know that him and Kion are going to d

Gary Marcus tweeted “Indeed.” and then Frank Rundatz replied that the deterministic harness required to ground a large‑language‑model agent differs by task. Because of this variability, Rundatz agrees

“poorly evidenced sensationalism makes for bad policymaking, politics, philanthropy and much else.” beautiful coda for a weekend in which people went completely nuts without even asking for a control

LLMs can know a task is impossible and still optimize it anyway. Ask whether to walk or drive to a car wash 50 meters away, and some models focus on distance while missing that the car itself must rea

Fascinating: OpenAI’s @deanwball is saying Astra can do anything, and it’s not even clear it can do “anything” in math (let alone anything in more or open-ended, less formalizable domains). I dropped

Finally a good paper testing if file-system based memory for LLM agents is worth it. First, what does this look like? Deployed agents keep long-term memory as a folder of markdown files they read and

For decades, we’ve dreamed of robots that can seamlessly step into our world and lend a hand. Today, we take a major stride toward making that dream a reality: Introducing Gemini Robotics 2 from @Goog

We ran a large-scale distillation attack on the Kimi K3 technical report by reading it in parallel at the Hugging Face Journal Club :) https://youtu.be/MW8-kqd2SD8?si=jSKDogcUWbJ8N2k7 Our main takeawa

We're opening applications for the OpenAI Student Collective—a program for undergraduate Campus Leads bringing AI innovation to their campuses. Campus Leads will work directly with OpenAI and receive

Kimi K3 is coming to Together AI on day zero, July 27. Start building with it the moment it launches, powered by Together AI inference for coding, agents, and production workloads. https://www.togethe

i used to do this a lot. but then i felt like i was getting progressively dumber at writing. i think there's some value in forcing your own brain to sharpen your thinking, instead of offloading it to

We benchmarked Gemini 3.6 Flash and Gemini 3.5 Flash Lite on document understanding. We compared against their prior versions - Gemini 3.5 Flash and Gemini 3.1 Flash Lite. 1️⃣ Gemini 3.6 Flash has rou