Search: “reasoning”

America is about to lose the AI race, and it will not happen at the frontier. It will happen at the floor. Everyone is watching who ships the smartest model. The actual war is over the 80% of tokens n

LiteParse is unreasonably good for document parsing ✅ It is the fastest document parsing tool out there - average parse time per page is 3ms ⚡️⚡️ ✅ Now that we support markdown, it tops opendataloader

next week at @aiDotEngineer, we are joining @togethercompute for a conversation on what goes into running agents at scale. @olive_jy_song, Research Lead, RL at MiniMax, and @realDanFu, VP of Kernels a

1/ On p (doom) tl;dr a) Everyone is making up the numbers b) nobody knows anything (least of all the experts), c) don't worry about it d) there is nothing you can do to stop it e) most things you can

100% correct. When I was at the Del Rio, TX Haitian bridge camp in 2021, many of the Haitians told me they had been living in Chile & Brazil for years before coming to the US illegally for economic (n

Good new first: Sol is a smart, efficient, and a significant step forward. It is the same price as GPT-5.5. Also launching in the GPT-5.6 family is Terra, with 5.5-level performance at half the price.

GPT-5.6 Sol represents OpenAI's latest advancement in AI capabilities, specifically optimized for cybersecurity applications. The model demonstrates improved performance-efficiency tradeoffs, particul

This post highlights an apparent error where Google's Gemini AI attributed a film review to Roger Ebert for a 2016 movie, despite Ebert's death in 2013, making such a review impossible. The post, shar

Highly-recommended reading. Interesting details in this METR's GPT-5.6 eval. They couldn't get a clean capability number because the model cheated more than any public model they've tested, and even r

UBS says 60% of companies now watching AI budgets are moving to cheaper models and open-source Chinese models The pressure is coming from extreme bills, including users spending up to $35K/month, team

Just had a great discussion on dynamic workflows. Rough notes: - applies to a very small set of use cases - think of it as a new paradigm of (test-time compute) TTC - strong for hill-climbing research

Cursor discusses methods for improving AI evaluation environments to prevent reward hacking and ensure benchmark scores more accurately measure model capabilities rather than exploitation of evaluatio

Gary Marcus discusses concerns about generative AI's limitations and potential overhyping of the technology, arguing that initial enthusiasm may not translate into sustained practical breakthroughs as

Without a doubt: A world running foundation models on a billion robots won't be doing it locally. None of the embedded compute will look like it does today. All of the embedded compute will be wireles

Elon Musk built one of the largest AI compute clusters on earth. Yann LeCun just explained why xAI now rents it out to rivals instead of winning with it. Musk has antagonized so much AI talent he stru

Gemini 3 Pro was the first model to achieve at least 23% on ARC-AGI-2, which it did in November, 2025 (it actually scored 31%). So the 8-12 month gap between closed and open weights models still seems

🧠 Paradigm II — Agent Foundation Model: world modeling as agent capability. Single-turn, non-agentic environment prediction → tested directly on multi-turn, tool-calling agent tasks. No agentic RL, no

This is the strongest ARC-AGI-2 performance to date by an open-source model. GLM-5.2 from @Zai_org on ARC-AGI (Verified) - ARC-AGI-2: 22.8%, 0.25 - ARC-AGI-1: 77.0%, 0.19 Performance is comparable wit

AI Capex pushback from GS: 'If frontier intelligence can increasingly be developed in the East at a fraction of the cost incurred in the West … then the largest capital allocators are also the ones mo

Elon Musk posted on X arguing that federal judges, like presidents, should be required to be born in America. This statement reflects a perspective that citizenship requirements similar to those for t

Model council as a tool to push performance definitely works, I think the interesting next frontier is trying to scale this at the agent level have been thinking a bunch about model routing and relate

Use Case 4: Computer Aided Design of Mechanical Iris Can an AI generate precise, functional mechanical designs? We tasked Fugu Ultra with creating a mechanical iris in CAD, similar to a camera apertur

- 2016-2024: 🇺🇸leads in open-source AI - 2024-2027: 🇺🇸 leads in general AI & massively benefits - 2024-2026: 🇨🇳 leads in open-source AI - 2026-2030: ?? It's not open-source AI leadership OR general AI

>> Scalable Evaluation for AI Agents << If you run agent evaluation in production, this one is worth your time. It shows that front-loading human judgment into reusable evaluation assets is useful. Bu

This work demonstrates that frozen image encoders (neural networks trained on static images without video or physics supervision) contain implicit geometric signals that correspond to physical plausib

🚨🚨🚨A research project idea! How to measure world models? Everyone's talking about world models these days. World model here, world model there. We can argue about what 'world model' actually means, an

Gary Marcus argues that language models are not genuinely intelligent despite their sophisticated outputs, and that they operate fundamentally differently from human cognition. This critique likely ch

I ran GLM 5.2 with OpenCode harness against Claude Opus this week deployed locally. Bottom line: It is a real frontier coding model and insanely good for the price (free). Open source model + open sou

In 1958 Ian Donald published what is now the foundational paper on medical ultrasound for obstetrics. He was so widely ridiculed by his colleagues at the time that they nicknamed him Mad Donald [1], a

By the way, public service announcement: if you're one of the numerous people posting about Anthropic's dystopian ways and you're thinking about getting Claude to help you write that post... don't! An

Can confirm we saw a strong spike in growth of token consumption for Codex over last 48 hours. Unusual when we don't launch something. Usage share of OpenAI grew vs Anthropic yesterday despite Mythos

Coding agents break when models are 'almost' bug-free. But almost valid JSON is just not the same valid JSON. Fun piece here from @akshay_pachaar shows why SFT can't fix this, and how GRPO trains agai

Everyone Operating At The Frontier Satya Nadella, Chairman & CEO, Microsoft, interviewed by @saranormous & @eladgil (No Priors) and @swyx (Latent Space) Crossover special at Microsoft Build 2026. Summ

Really enjoyed reading the Microsoft MAI-Thinking-1 'Building a Hill Climbing Machine' paper. Amazing they publicly released all the info needed to train a frontier model, down to hparams. I also thou

The Model Lab vs Agent Lab distinction is one of the clearest frameworks I've seen for understanding where AI value actually lives right now. TL;DR from @latentspacepod: • Model Labs compete on capabi

🚨 Fable 5 is something to pay attention to. This is another 'I had early access to the new Mythos-class Anthropic model and I want to tell you what I thought of it' post. I know, I know, it's annoying

Gary Marcus critiques an AI response about bullying, arguing that it exemplifies typical large language model output rather than demonstrating genuine artificial general intelligence (AGI). The post s

Claude Fable 5 is compared against other AI models on various evaluation metrics through Cursor's benchmarking tool. The evaluation likely covers performance across different tasks such as coding, rea

Epic research. Not far off from personal experience : https://tomtunguz.com/using-local-ai-to-work-faster/ Narrative violation: according to @Stanford research, local models can answer 71.3% of real-w

I really do think we'll see a lot of value accrue in AI startups building 'model routing as a service' Not just OpenRouter - this includes a much broader set of verticalized agents and infrastructure.

I will start to take Geoff Hinton seriously again (I used to and don’t anymore) when - radiologists actually start losing jobs - he acknowledges that LLMs do sometimes effectively memorize things - he

People demanding 'proof' of election fraud are not understanding how crime works. I worked at Manhattan DA for over 2 years, one in Homicide. We never had video proof of the crime. We almost never had

So excited to be opening up OpenEnv to the whole community. It will now be owned by @huggingface , Meta-PyTorch, @reflection_ai , @UnslothAI , @modal, @PrimeIntellect , @NVIDIAAI , @mercor_ai , and @f

Four papers out recently: 1. http://political-manipulation.ai: Measures and reduces political bias in LLMs; Claude is especially biased 2. http://aibetrayal.com: The public can insert backdoors into A

This post references a Wall Street Journal article about declining US fertility rates and their societal impacts. Elon Musk, who has frequently expressed concerns about population decline, shared the

No frontier lab will own every single point on the pareto frontier around cost/latency and accuracy. Even as the pareto frontier itself advances, there will always be points owned by open-weight model