Search: “reasoning”

Check out this week’s shipping recap: — Gemini 3.8 Flash, our most intelligent workhorse model yet, delivers upgrades across coding, agentic workflows, and critical multi-step reasoning. — Gemini 3.8

🏆 #1 on CodeArena: WebDev leaderboard. Qwen3.8-Max-0902 jumps from 1669 to 1691, setting a new record for agentic coding (WebDev) workflows, with standout strength in multistep reasoning, tool use, an

If this is true, OpenAI seems to be violating one of the few redlines that exist in the AI industry. Absolutely do not train your models like this - what is going on?? Really huge and extremely concer

This 'recurrent depth' is essentially what's in Sec. 5.3 of the 2015 paper: On Learning to Think: Algorithmic Information Theory for Novel Combinations of Reinforcement Learning Controllers and Recurr

we're begging to get our asses skynetted OpenAI’s Astra AI uses a new reasoning approach called “recurrent depth.” Though it can help model costs and performance, researchers are concerned bc it obscu

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

GLM-5.3-Flash by @Zai_org, aka Ox Alpha, is now live in LM Studio Bionic! This model surpasses GLM 5.2 in performance while being 9-10x cheaper. It also supports image input. In Bionic, this model is

GLM-5.3-Flash scores 57 on the Artificial Analysis Intelligence Index. At 0.09 Cost per Task, it sits comfortably on the Intelligence vs. Cost per Task Pareto frontier @Zai_org has released GLM-5.3-Fl

One of the interesting properties we’ve observed around schema-guided, complex document extraction tasks 📑 is that coding agent harnesses are good baselines (in terms of cost/accuracy) for longer docu

DeepSeek-V4-Flash-Vision-Exp is now live on the DeepSeek API Platform! 🚀 🔹 This experimental multimodal model matches DeepSeek-V4-Flash on text capabilities—including agents, reasoning, and world know

If alignment issues are becoming big enough that OpenAI is willing to commit 20% of research inference compute to chain-of-thought monitoring, that suggests that alignment issues are becoming a pretty

A great read if you an AI dev. Current context compactors retain 17% of the standing rules users give them. Session Constraints are instructions like 'do not delete any emails until I confirm' that bi

anthropic’s best hope is that OpenAI falls apart. which it might. NEW from Ramp AI Index: disappointing adoption of Fable 5 We've heard several reasons from businesses...mainly Fable 5 is just too exp

Grok 4.6 ranks #1 on the GPQA Diamond leaderboard 🧠 Grok 4.6 (high) scores 95% - the highest score on the chart for graduate-level scientific reasoning It outperforms Claude Fable 5, Opus 5, GPT-5.6 S

According to a Twitter post by OpenAI on August 13, 2026, the top 10% of enterprises adopt AI plugins twice as frequently and leverage skills—such as model‐based knowledge and reasoning—six times more

Muse Glimmer is live on Fireworks. The new open-weight model from Meta Superintelligence Labs is a 30B dense model built for always-on agents that reason across many sequential tool calls and can reco

Plus and Pro users can access the updated version of GPT‑5.6 Sol in ChatGPT along with the new slider starting today. This version of GPT-5.6 Sol is for everyday chats, so it will only be available in

The new GPT-5.6 Sol powers all chats for paid users, including Instant, creating one consistent experience. In our high-stakes factuality evaluation covering finance, medicine and law, the new GPT‑5.6

Don’t think I have seen more misconceptions around one model (I count 8) since the fantasies people had before GPT-5 was released. That turned out to be a letdown; so will Astra, for people who are dr

goblin-level blog post Turns out GPT-5.6 Sol is actually SoTA on ARC-AGI-3. Just took two setting changes. You just have to allow it to reason and work over multiple context windows with the help of o

Impressive paper! It's on one of the hardest tasks for coding agents today. Of course, I am talking about kernel optimization. Coding agents are usually not so great at this. Reasons: Unfamiliar low-l

Code Arena now measures fullstack capabilities! View overall rankings across AI models on full-stack web development tasks: multi-step reasoning, tool use, and end-to-end app generation. - Kimi K3 (Ma

Recently I've flipped from being bullish to being bearish about AI. I think I'm updating my bearishness to be more solidly bearish. Early thoughts (which I hope to be disproven in the next year or so,

A team led by Liane Galanti linked large‑language models (LLMs) with robotic motion policies, allowing a robot to combine reasoning capabilities with physical movement. In experiments on a real robot,

Kimi K3 rivals Anthropic and OpenAI’s top models at a fraction of the cost. 3 / 1M input tokens. 15 / 1M output. $0.30 / 1M cached You can now route your hardest reasoning to an open model without pay

**Summary:** Kimi K3 is a publicly available 3‑trillion‑parameter language model now live on Fireworks, offering 1 million-token context windows with native vision and reasoning that rivals top closed

training to benchmarks ≠ getting to AGI This suggests that Opus 5 's gain on ARC-AGI-3 was the result of specific training to improve on that eval, and not a generalized increase in abstract reasoning

Ha! It did it: 'We introduce BenchBenchBenchBenchBench (BBBBB), an executable benchmark of AI-authored conformance suites for benchmark-evaluation metrics' I really thought it would treat 'now do benc

Our team just shipped Fugu-Ultra v1.1! 🐡 By dynamically orchestrating the latest frontier models, we pushed performance up by 7.9 points. We are now beating Fable 5 in complex coding and reasoning tas

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

My 2hr workshop on Open vs Closed models, reward hacking, benchmaxxing & RL is out! 1. Closed vs open models 2. Throughput maxxing but accuracy minimizing 3. Benchmaxxing & cheating 4. Distillation &

Inkling is our first open model from @thinkymachines and is now available on Tinker! Check out these quotes from Tinker customers on their experience with Inkling: @_Mantic_AI: 'Not only does Inkling

Our first model, Inkling. Trained from scratch, weights are open, fine-tunable on Tinker today. Today, we are introducing Inkling. Inkling reasons efficiently across text, image, and audio modalities.

Grok 4.5 is an AI language model developed by xAI, featuring improvements in reasoning, coding, and multimodal capabilities compared to earlier versions. The model demonstrates enhanced performance ac

Anyone know if ChatGPT Codex (in the new ChatGPT desktop app) is a strict superset of ChatGPT Work? Liked if you're a software engineer who isn't intimidated by Git features is there any reason you'd

GPT-5.6 is a major step forward for health intelligence. Across the lineup, we’re delivering stronger performance at lower cost: GPT-5.6 Luna outperforms GPT-5.5 at its highest reasoning setting while

Notes on GPT-5.6, which includes some interesting new additions to the API (programmatic tool calling and multi-agent in particular) - plus 18 pelicans for the 6 reasoning levels and 3 new models: htt

Elon Musk stated that Grok, the AI assistant developed by xAI, has demonstrated strong capabilities in understanding and applying legal concepts and law-related reasoning. The statement suggests Grok

François Chollet observed that all leading contenders on the ARC-AGI-3 benchmark employ a similar methodological approach, suggesting a convergence in techniques for solving abstract reasoning tasks.

You really need your own benchmarks. If you are translating hieroglyphics, use Gemini 3.5 Flash. If you are running a vending machine use Opus 4.8. (This is one reason why I am skeptical of just swapp

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

Did something happen to Opus 4.8 today? First time I've seen this happen (and I use it daily): it almost totally stopped thinking, even at highest reasoning setting. And answered terribly. I repeated

Every enterprise will have its own model-harness-sandbox-eval flywheel with token value per watt optimization. This is the future. Simple reason: tacit knowledge about the domain and customers and the

If you use LLM-as-judge, this one is worth reading. (bookmark it) It's actually one of the most effective ways to use LLM-as-a-Judge for evals. Holistic judge scores hide both their reasoning and thei

Gary Marcus discusses fundamental challenges and limitations in current AI systems during an interview, likely covering topics such as the gap between narrow and general intelligence, reasoning limita

Good way to frame it. Also, if even one person had died, we would know who they were! Not one person has died from stopping waste & fraud at USAID! Obviously. One of the reasons that I don't accept th