Search: “context”

If Leopold had read this on June 26 and trimmed his bets accordingly, SALP would not have melted down. I laid everything out. https://open.substack.com/pub/garymarcus/p/the-month-generative-ai-lost-it

After a few more hours, I think I've figured out Opus 5. Opus 5 is trained to be more agentic than anything I've used. All Claude 5 models are like that. So what changes? The way to interact with Opus

AI can write more code than any team can review by hand, and a pull request can look fine while hiding a security issue or missing a requirement. In our new short course, AI Code Review, built in coll

Coding agents are helping scientists spend more time advancing research, taking on everything from routine maintenance and targeted optimization to complete redesigns and new systems. While agents can

Very interesting paper on LLM reasoning. They find that frontier models can exhibit invisible reasoning by leveraging semantically irrelevant filler tokens. In other words, invisible reasoning can ser

Powered by Cohere’s secure agentic platform North, Automations enable employees, regardless of their technical abilities, to: 1. Turn complex workflows into simple outputs 2. Stay in control at each s

This release marks a significant step forward in making agentic AI practical, scalable, and deeply integrated into daily business operations. Automations is available today to all North customers. Lea

A new model launch is not a product update ‼ It's one of three things, and you don't know which until you test it. Sometimes it's nothing: the model improved inside the same distribution, your harness

We're parsing some of the hardest financial documents into clean, plaintext/structured outputs through a live webinar. Come check it out! https://watch.getcontrast.io/register/llamaindex-from-complex-

Vibe coding isn't just for your PC anymore. Here's LLM Hub generating and previewing code entirely on an iPhone. Everything runs locally. Using the RunAnywhere SDK, Timmy Qian ported LLM Hub, an Andro

Interesting finding on frontier models. It turns out that frontier models can write proofs but stumble on faithfully copying a long block of text. This has huge implications. It sounds trivial, which

Blog and paper: https://research.nvidia.com/labs/gear/robottt/ Presented to you from NVIDIA GEAR Lab. Please check out Yunfan's technical deep dive! https://x.com/YunfanJiang/status/207740099214934055

Excited for our first general model Inkling -- open weights, 975B, natively multimodal (text, image, audio). Available on Tinker, HuggingFace and partners. It is yours to personalize and use openly. I

🆕 5 Trends That Defined AI Engineering at World’s Fair 2026 https://latent.space/p/aiewf26trends @ricmac's big recap of @aidotengineer: 1. The focus shifts from agents to systems 2. Loop engineering i

congrats on the launch! As the task horizons on agents gets longer, we need more work evaluating and training models to be better in long-running, open-ended, evolving real-world environments. Today w

Ever wanted to quickly turn a PDF into clean text to paste into your favorite AI agent, without having use CLIs or open the browser? We built exactly that. Using @TauriAp ps, with a Rust backend power

New model for AMD Strix Halo users: My 198B Step 3.7 Flash release was a big hit, but this one may be even better: 298B-parameter Hy3, now running on a new 2-bit FPX codebook designed to map efficient

This post likely reflects on the distinct emotional or relational significance of being called 'daddy' by a bisexual male partner, suggesting that the experience carries different meaning or impact co

During a Bloomberg interview, Yann LeCun (@ylecun ) explains why LLMs are limited in terms of real-world intelligence during a Bloomberg interview. 'Language is a very approximate, reduced, quantized,

If there is ANYTHING I have learned in my early Fable 5 testing, it’s that the judgement calls from AI are getting better and better. Not perfect and not (yet?) a substitute for a human’s taste, conte

Come and chat with us in ICML 🥳 Excited to present Temporal Straightening for Latent Planning in the Tuesday morning session #1509. Let’s talk about world models, JEPA and representation learning. Age

Multimodal prompting enhances coding agents by enabling them to process and integrate multiple types of input data—such as text, images, and code snippets—simultaneously, improving their ability to un

Don't train the model, evolve the harness. I read a brilliant blog post from Hugging Face where they took a frozen open model scoring 0% on a hard legal agent benchmark, left its weights alone, and le

How to keep AI spend flat while token usage grows exponentially: Not with friction and spend alerts. With better defaults, routing, and caching. Better Defaults (not Usage Caps) – Engineers can choose

What is actually limiting model routers for coding tasks? Most routers treat picking a model as a static, one-off classification. This paper identifies the real bottleneck as information deficit. Simp

One thing we’ve seen pretty consistently: shipping the first version is only a small part of the work A key part of building reliable agents is having a repeatable lifecycle for improving them over ti

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

Years ago someone asked me what comes after the AI-first enterprise. I said the autonomous enterprise. The core of it is a flywheel: four parts (well, five including feedback) that feed each other, an

🚨 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

What are loops, and how do you build one? A 'loop' is the repeated process where some event or input kicks off an action. For example: 1. CI fails -> you fix it 2. CI fails again -> you fix it 3. CI p

We're building this at LangChain Fleet lets you create and manage a fleet of agents. Each agent specializes in a workflow, e.g. inbox management, blog writing, competitor research, candidate recruitin

Before the week ends, let's acknowledge one of the most INSANE week ever for open AI, with 25+ notable open-weight drops across every modality: 🧠 LLMs → NVIDIA Nemotron 3 Ultra: 550B hybrid Mamba-MoE,

THE PROMPT THAT REPLACES YOUR WORKFLOW AUDIT: You are my business transformation strategist. I want you to find the real AI-era business transformation hiding inside my work by starting from goals, bu

We wrapped a live session on M3 yesterday with the @togethercompute team & our researchers @zpysky1125 and @HaohaiSun A few highlights 🧵 1. MSA (MiniMax Sparse Attention) is the star ⭐️. Unlike CSA/HC

6 months ago I said I won't stop until I have Kimi at home, after 10+ botched REAPs I finally have it Needs benchmarking of course. - 45 tok/s decode - 954 tok/s prefill no cache - 95k+ tok/s cached p

activegraph longmemeval experiment #1: https://x.com/ActiveGraphAI/status/2059106393433780292?s=20 [technical blog post] “Evidence Compilation Before Semantic Memory: ActiveGraph on LongMemEval-S” ——

This recent @latentspacepod pod was a good one (they usually all are). One specific piece resonated: They were talking about AI agents writing code, and the line was basically that without caution / d

Model-Harness-Task fit! it’s clear that RL post-training produces a model-harness fit via tool shapes and prompting as models are trained with the harness in the loop. Mentioned this in a previous Lan

Activegraph's website, newsletter, and marketing are all run on Cofounder! [technical blog post] “Evidence Compilation Before Semantic Memory: ActiveGraph on LongMemEval-S” —— 🔍85.6% QA accuracy and 8

// Language Models Need Sleep // Let your agents 'sleep', folks. On a serious note, this is a fascinating paper on getting the most from long-horizon agents. Here is the problem with agents today: Att

ran my first benchmark this weekend (longmemeval) mostly to test activegraph, learned a lot! - this is a stepping stone to show the event based agent system works. the AI convinced me not to start wit

System scaling is the next real bottleneck in agentic AI. If you build agent orchestration layers, this is a clean map of where the engineering leverage actually sits. The labs own the model. You own

fun read on real tradeoffs & design decisions we debated when designing Engine for the data scale that customers produce one common thread is that we’re pretty strong supporters of just giving the age

Google’s new Gemini 3.5 Flash is the clear leader on the Intelligence vs Speed Pareto frontier and makes large gains on GDPval-AA (real-world agentic tasks), but is 5x the cost of Gemini 3 Flash @Goog

okay this email based mission control i prototypes last night is pretty cool: > starts with email > grab and summarize content about participants from crm/notes > auto bucket emails into workstreams u

AI has a secondary intent problem. I was negotiating a rent renewal (I live in NYC and they raised rent 10% because I guess they were bored). The property manager said she would 'do everything she cou