All entries

APEX-Agents now has a @huggingface leaderboard for open-source models. APEX-Agents is our frontier benchmark for whether models can do the real work of consultants, lawyers, and bankers. https://huggi

ASI* is GPT-X deciding I want to throw a party for myself and every human on earth will get a personalized email them that will help them in some way, and also here is a cure for a bunch of diseases *

At @sequoia’s AI Ascent last week, @gdb told me something that stuck: in late 2024, AI wrote ~20% of @OpenAI's code. That number is now 80%. We also got into why human attention, not compute, is the r

> be me > 'the internet is polluted by ai slop, we need low-background tokens' > 'wouldnt it be cool if we could time travel and see what our ancestors 100 years ago would say to us' > all the existin

Big Tech companies are projected to spend approximately $700 billion on AI infrastructure and development in the current year, a figure that AI researcher and entrepreneur Gary Marcus has criticized a

OpenAI released a significant update to Codex, expanding its capabilities beyond programming tasks to handle general computer work and automation. The upgrade was announced by Sam Altman on X (formerl

Building a document processing pipeline at scale is hard, and is one of the reasons that it's hard to DIY your own document OCR solution by relying on LLM APIs. Your orchestration pipeline needs to ha

Building scalable, distributed document processing pipelines isn’t easy. That’s why we teamed up with @render to build a system that: 📝 Leverages the LlamaParse platform to parse, classify, extract, a

By the way, if you want to work with me but don't think the above role is the *perfect* fit, I'm also hiring 4 other roles: > Head of Content > Head of Sales > Ad Creative Strategist > Branded Content

arXiv:2604.26269v1 Announce Type: cross Abstract: The essence of good creative writing is calibrated surprise: when constraints from all relevant dimensions act together, the feasible solution space c

Can AI dance and choreograph? Yes. Yes, it can. I had a song stuck in my head last morning, and before noon, I managed to get it out by making this music video. From idea to execution, it was created

LangChain's DeepAgents framework is receiving frequent updates and deployments, with documentation available at their official docs site. The project maintains an active development cycle with updates

Claude Code is tuned for Claude. Codex is tuned for OpenAI models. Until now, 𝚍𝚎𝚎𝚙𝚊𝚐𝚎𝚗𝚝𝚜 had fixed harness defaults, which meant it couldn't take advantage of the provider-specific optimizations that

Claude's Computer Use capability has been integrated into the Codex app, delivering a 42% performance improvement for this use case in the latest update. This enhancement likely demonstrates improved

Elon Musk posted a congratulatory message to the SpaceX Falcon team, likely celebrating a successful mission or achievement related to the Falcon rocket program. Without access to the specific post co

This post discusses the liability and accountability questions that arise when problems occur in AI-automated workplaces, exploring who bears responsibility—whether the AI developer, the employer, the

Cursor Security Review is now available for Teams and Enterprise plans. Run two types of always-on agents: 1. Security Reviewer checks every PR for vulnerabilities and leaves comments. 2. Vulnerabilit

Cursor offers customizable security agents that allow teams to configure triggers, add custom instructions, and tailor agent behavior to match their specific security requirements and workflows. This

Dear @elonmusk, If you still genuinely care about AI safety, you can’t let the Trump administration leave the AI industry almost entirely unregulated. You just can’t. - Gary The judge just instructed

Deep Agents is a framework or library developed by LangChain that enables the creation of sophisticated AI agents with enhanced reasoning capabilities. The project likely provides tools and abstractio

🚀DeepAgents deploy is a simple, configuration driven way to get an agent harness deployed to the cloud deepagents.toml is the file that configures it. It has four sections: - agent - sandbox - auth -f

Demis says he wants to see a Western open source AI stack and that we’re losing to China. He also says Google doesn’t have enough compute to build two frontier (open and closed) models, which is why G

developers talk about agent harnesses/orchestration like it's back-end architecture and that if they could only get it right, they can keep building on top of it but what if it's more like personal/or

Distillation is largely an industry standard and not just something done by Chinese labs targeting OpenAI/Anthropic. Many American companies also distill Chinese (open) models. LIVE TRIAL UPDATE: Open

does anyone else remember how Jacob Devlin went to the media about how Google was supposedly distilling from ShareGPT (it was not) and the resulting blowback? How times have changed 😅 https://futurism

dropbox has dropped the ball. how is this a 6 billion dollar saas. ultra slow navigation, doublecharges for storage (of course, extremely happy to help you buy more storage, really shitty tools to hel

Every year for AI Ascent, @gradypb, @Konstantine and I get to share some perspectives on AI and where things are headed. This year's talk was about the arrival of agents and the race to deploy them ac

Evinova, a separate health tech company within the AstraZeneca group, builds software that modern clinical trials run on. They put Devin to work on regulatory documentation, bug triage, tech stack mig

Exciting direction - can see immediate gains for many teams trying to prevent lockins for cost/legal reasons - my mental model on this is shaped by the meta-harness paper that we need to think of harn

Falkor's new GraphRAG SDK just took #1 on GraphRAG bench and yes it's open source 😆 Token costs spiking. Responses too slow. Users don't trust the answers. All three are retrieval problems. GraphRAG S

Finally this theme is taking off 👇 For the record : it’s not only a matter of cost. This applies both to agent design (particularly when testing more innovative, substrate-first archis...) and to issu

Adaptive Data is a Together AI tool designed to help teams improve the quality of datasets before fine-tuning language models by enabling analysis, adaptation, and refinement of training data. The pla

Fireside chat at Sequoia Ascent 2026 from a ~week ago. Some highlights: The first theme I tried to push on is that LLMs are about a lot more than just speeding up what existed before (e.g. coding). Th

For better or worse, regulation for closed-source models served by a few (quite large) companies is easy. It is not as easy to imagine how you regulate open-source models that can be served by a range

Forget goblins, things that GPT-5.5 really likes in its fiction: lighthouses, the ocean, maps, bells, clock towers with bells that ring impossible times, Mira Vale, resonances and echoes (Claude and G

Frameworks are dead. @jerryjliu0 of LlamaIndex called it — and then rebuilt the company around what actually matters: context. New episode of Beyond the Pilot: https://www.youtube.com/watch?v=HbXvX-Kt

Codex allows users to iteratively refine documents and presentations by opening files, requesting modifications, and making adjustments within a single conversation thread. The tool enables a collabor

[frontend] UI/UX is a surprisingly big part of building agents! It takes a while to hook up streaming, thread history, all the endpoints We do it for you! Get a frontend you can expose to users out of

Fun fact - if you have a recent commit that mentions OpenClaw in a json blob, Claude Code will either refuse your request or bill you extra money. This is an empty repo, I'm just calling Claude Code d

Generate multiple character angles from a single image in one click. Upload → Run → get different views automatically with Qwen Edit. Edit prompts if you want more control, or keep it fully automated.

Goblin and related magical mentions were overrewarded in training, and the behavior was reinforced over successive models. We removed the goblin-affine reward signal for future models, and filtered tr

GPT-5.5 is going to have a party for itself. it chose 5/5 at 5:55 pm for the date and time. if you'd like to come, let us know here: https://luma.com/5.5 codex will help the team pick people from the

Clem Delangue, CEO of Hugging Face, announced that his company was recognized by TIME magazine as one of the 10 most influential AI companies of 2026. The post emphasizes Hugging Face's commitment to

Grok 4.3 beta just went insane with front-end design I gave it one single prompt: “Build a WWE-style Fight Club website with 10 fighters” It designed a stunning, premium-quality website that genuinely

Grok Imagine Agent Mode (Beta) just went live on Grok web It’s a full creative agent working on one infinite open canvas Grok Agent plans → generates → edits → iterates everything automatically in the

Hackathon info https://x.com/NousResearch/status/2045225469088326039 The Hermes Agent Creative Hackathon starts now 16 Days, $25k in Prizes Presented by @Kimi_Moonshot & @NousResearch For the tinkerer