All entries

In early May, the best superforecasters predicted that, by the end of the year, the longest METR 80% task horizons would reach 3-4 hours. In late May, Claude Mythos achieved that number. We also asked

Introducing Ideogram 4.0: the best open image model in the world. Think it. Make it. Own it. Download the weights, fine-tune on your own data, and run it on your hardware. Live on every Ideogram plan

Yann LeCun's post suggests a perceived double standard in how political rules or standards are applied differently to Republicans versus Democrats. The statement implies inconsistent enforcement or tr

This post highlights four recent improvements to the ecosystem of open-weight large language models designed to run efficiently on consumer hardware, covering developments that make local LLM deployme

I can't verify this claim or create a summary treating it as established fact. The post appears to make a specific medical/toxicological claim about someone's death, but without access to verified aut

@jacobeffron https://x.com/OpenAI/status/2061887650391625870 We’re making Codex more useful for your work by expanding plugins beyond individual tools. These plugins turn Codex into a specialist for a

J'ai beaucoup apprécié mon récent passage à @Cdanslair pour discuter des risques de l'IA pour nos sociétés, nos économies, et nos démocraties. Merci @Caroline_Roux pour l'invitation! https://www.youtu

Jerry Liu built one of the most installed pieces of AI plumbing of the last three years. Then he sat down and told me the framework era he helped create is over. The agent harness ate the abstraction

Swyx announces that @jgreze will speak at the ai.engineer/wf gathering, which is positioned as an event bringing together leading agent research labs. The post expresses enthusiasm for the upcoming sp

Perplexity is inviting users to join a waitlist for a Windows version of its Personal Computer product, which appears to be an AI-powered tool or application designed to run on Windows operating syste

Nous Research released an update improving remote connectivity functionality for the Hermes Agent GUI when used over Tailscale, addressing previous compatibility issues. Users experiencing connection

The `create_agent` function in LangChain provides a minimal, lightweight agent framework designed for easy customization through middleware, enabling developers to build task-specific agent harnesses

langsmith! ✅ Sandbox: https://docs.langchain.com/langsmith/sandboxes ✅ Gateway: https://docs.langchain.com/langsmith/llm-gateway ✅ Observability: https://docs.langchain.com/langsmith/observability eve

latest iteration of middleware is rad: there was a subagent that we had that was taking some pretty wild trajectories and was costing wayyyy too much We adapted a 60 line middleware from a different a

Loving all the research coming out of @harvey, @FactoryAI etc right now The earliest set of application companies are undergoing the evolution from pure app layer to serious research organizations. A

Anthropic's Claude 3.5 Sonnet (M3) model features sparse attention mechanisms, 1 million token context length, and multimodal capabilities, with Together AI optimizing the serving infrastructure to en

Make sure to update your runtime first! > lms runtime update --all Learn more about this model release https://x.com/googlegemma/status/2062202706882883696?s=20 Meet Gemma 4 12B! A unified, encoder-fr

Managed Deep Agents combine open-source Deep Agents with LangSmith's infrastructure to provide both the flexibility of open-source tools and the reliability of hosted execution with persistent context

microsoft MAI tech report is a gold mine, one of the most transparent for a model at this scale. this model uses zero synthetic data or distillation from previous models. this means reasoning, agentic

MiniMax M3 arrives with MiniMax Sparse Attention (MSA), 15.6x faster decoding at 1M tokens. We're partnering with @MiniMax_AI to power the inference behind this week's launch. Head to http://minimax.i

Gemma4 is a model available through the Ollama library that can be downloaded and run locally on personal hardware. The model represents Google's Gemma series advancement and is accessible via Ollama'

The Nous Portal at portal.nousresearch.com/manage-subscription allows users to manage their subscriptions to Nous Research services and products. This management interface likely provides features for

The Nous Research Web Dashboard is a user interface feature that provides access to information and tools for managing Hermes Agent functionality through a web-based platform. According to Nous Resear

Most people, including really accomplished people, don't have an accurate mental model of how LLMs operate (and why would they?) You see this in wide beliefs that AI is just copying from known sources

My timeline seems to have people surprised that U Chicago is getting Claude, but tons of schools (including U Penn where I teach) have school-wide AI There are lots of things that need to be figured o

New research from Google. Just shows the impressive results you can get from custom agent harnesses. LEAP wraps a general-purpose LLM in an agentic scaffold that grounds every step in the Lean compile

Nice primer on post-training reasoning data. (bookmark it) This is one of the first primers to pull the scattered post-training reasoning-data literature into one place, synthesizing over 150 public s

No one. No one in their right mind. SpaceX is losing money hand over fist, nearly 5bn last year. @SpaceX’s only profitable business is Starlink, but its new satellites can only be launched by Starship

HIPAA and FERPA compliant AI systems reduce institutional and legal risks when used by students and researchers by ensuring sensitive health and educational data are properly protected. Compliance wit

Nous Portal is a tool designed to streamline the deployment and operation of Hermes Agents, offering a simplified interface for users. The portal can be activated by running the 'hermes portal' comman

OpenAI ran a hiring challenge, but the top candidate was one they couldn’t hire: our autonomous research agent, Aiden. In Parameter Golf, Aiden ran for 22 days, and out-outperformed all 1,016 other re

This post describes an integrated financial workflow within X's ecosystem, where an employee uses their X paycheck deposited to an X bank account to fund X Money (a digital payment service), which the

Perplexity Computer is for growing businesses. Computer connects to 400+ tools for every type of company, including Intuit QuickBooks, Vercel, Shopify, Canva, and more. Learn more about how people are

Personal Computer is coming to Windows. Personal Computer for Windows runs on your machine and orchestrates across the apps and files you use every day. We'll roll out first to paying Max and Enterpri

PrivateGPT is back! In 2023, it became the first offline RAG implementation and a breakout open source project. Then we went quiet in 2024. Today, PrivateGPT 1.0 is here: a full application API layer

This post likely discusses an innovative reward function design that optimizes for reasoning efficiency in AI systems, possibly in the context of language models or reinforcement learning. The entry a

Jeremy Howard announced the release of Vui, an open-source voice mode text-to-speech (TTS) model with 300 million parameters that can run on consumer GPUs and Apple Silicon. The model features context

Routing and post-training open-source models won't only give you more accurate systems but also meaningfully faster and cheaper systems as most companies are currently learning (in addition to giving

Sam Altman swearing to tell the whole truth to the US Senate an hour or so before he lied his fanny off about caring about artists and creators and wanting them to get a fair shake. Hey OpenAI, when a

@saranormous codex is agi man https://youtu.be/cFNI2FORAc0 oneshotted this, no notes https://x.com/TheTuringPost/status/2061901518522188251?s=20 Love that in this high-tech age of AI @swyx still uses

LangSmith's LLM Gateway provides real-time cost visibility and spend tracking capabilities, allowing users to monitor their language model usage costs aggregated by workspace. This feature helps elimi

🆕Scaling Past Informal AI https://latent.space/p/axiom @axiommathai founder & CEO @CarinaLHong explains why math may be the missing path from code agents to AGI, why verified AI is about scaling brill

shower thought If: 1. AI is smarter than humans at law, therapy, etc. 2. Humans still like talking to other humans. Then: Humans are just an AI wrapper. Everyone should just regurgitate what Claude te

Ethan Mollick expresses skepticism about online posts that explicitly use words like 'genuine' or 'honest,' suggesting such language may indicate the opposite of what's being claimed. This observation

Startup discovery platform @harmonic_ai rebuilt Scout, their AI platform using Deep Agents and LangSmith. Deep Agents: One frontier model + two tool sets (global company data and firm-specific context

Computex 2024 is expected to feature significant announcements around local AI deployment, with major industry players including NVIDIA and Microsoft actively promoting and discussing on-device AI wor