Knowledge catalogue
Anthropic says its first Fermat runs failed not on math, but because dozens of Claude agents lost track of shared project state. They only succeeded after moving to Prove2Me's DAG of statements and pr
Banger paper from Microsoft and Cornell. If you have looked at thinking tokens and decided you cannot afford the context, read this one. (bookmark it) A pause token buys the model extra compute for ea
Chamath says Nvidia's Hugging Face acquisition is as a massive move against AI becoming a closed source oligopoly 'Jensen and I were texting this morning. I said to him, this is the most incredible th
Easily pick up your codex or Claude code sessions in Hermes Have you ever been working in Claude Code or Codex and wished you could pick up this exact session in Hermes Agent? Now there's a single com
From GPT-4 to GPT-6. Sparks was a remarkably prescient paper that got a lot of pushback at the time, but absolutely sensed where the vibes were heading with LLMs based on a lot of qualitative experime
Full Friday Showcase replay. We went deep on Replit MCP with Amadeo Pellicce (@amapel) and Jean-Luc Thumm (@jeanlucthumm) from the Foundry team, the two who built it. What MCP actually is, why we open
GPT-6 is pretty good tbh @viktaur27 @Teslarati The rate of improvement from original GPT to GPT-3 is impressive. If this rate of improvement continues, GPT-5 or 6 could be indistinguishable from the s
How we think about the “wiki incident,” where our agents wrote to several internet sites: it’s past time for us to define standards for when and how we share misalignment incidents, not just misalignm
SemiAnalysis reported on April 17, 2024 that they had tested Anthropic’s Fable 5.1 alongside OpenAI’s Astra language models. According to their review, the comparison revealed a decisive advantage for
I do not fear the rise of artificial intelligence. But I do fear the rise of a small set of billionaires and companies who seek to build such things to increase their power and their wealth, all witho
Introducing off-peak hour token rates. DeepSeek-V4-Flash and Pro are now half price outside of 12:00 to 18:00 UTC on weekdays (5am-11am pacific), and all day on weekends! Off-peak pricing will be avai
Legit think he's stopped caring as much since it's apparent the media is never going to do any reporting on what the fuck is going on with it. oh my goodness -- Trump's hand looked absolutely awful ye
Listen to my conversation with @GaryMarcus about why he’s stated that OpenAI needs to Pause NOW. Gary’s not ready to pause all frontier AI, but Pause is a solution to many problems, including OpenAI’s
me: AI isn’t that smart & remains unreliable and now I am worried that monitoring-which likely could have prevented multiple incidents-is being reduced, and also worried that OpenAI has consistently n
New TIL on using Blender with coding agents on macOS: https://til.simonwillison.net/llms/blender-coding-agents-macos GPT-6 Astra (medium): > Use the already install /Applications/Blender to render a s
Note this is separate from privacy - you can make sure your data is not used for training by turning that off in OpenAI and Claude. You can also use an open weights model, but as they continue to impr
OpenAI quietly altered the performance metrics it reported for GPT-6-Astra in ways that favored the new model, and it continues to change others post-launch. Good reporting here from @EmilyForlini htt
Over 600 autonomous rocket landings so far Most people don’t realize how insanely autonomous Falcon 9 already is Once Falcon 9 lifts off, almost the entire mission is executed by the rocket itself And
Pause OpenAI - they are a deeply unethical company building technology that they are demonstrably not able to control. OpenAI quietly altered the performance metrics it reported for GPT-6-Astra in way
Really disappointing “Save College Sports Act” segment on GameDay. Not one dissenting opinion. No one speaking up for labor. Most watching don’t know the obvious conflicts of interests. They did a bet
so does this digital fly have the memory of the real fly that was scanned or what I’ve successfully run the full retained MaleCNS v1.0 fruit fly connectome, all 166,700 neurons, inside Minecraft, with
This is a weird behavior in coding models and something worth looking into. It turns that some models over-edit code that another models wrote. There is a high chance that your repo now has commits fr
This is very similar to how we think about context management in Deep Agents: compaction should reduce what is sent to the model, not erase the underlying work history. In Deep Agents, the logic is: -
We benchmarked GPT-6 Astra on hard document parsing and extraction tasks. It is the best frontier model for one-shotting document extraction over short and medium-sized documents: 📊 97.2% over short d
100%, which is the core of why I called for a pause on OpenAI in my newsletter today. The Hugging Face incident was probably just the tip of the iceberg. OpenAI has lost control and they're hiding imp
A group of incredibly talented and dedicated engineers made Cybercab happen. If you want to join them, we’re hiring AI engineers across the board: scaling, data, RL, reasoning, evals, world models, au
Agentic-SDLC requires continuous learning 🎙️ New episode of The Agentic Review is live. @itamar_mar and @nnennahacks sit down with Emilie Schario, VP of Engineering at @anacondainc, to dive into AI RO
Also in podcast format New episode: the gripping tale of what is happening in China’s AI companies right now. Are they coming for Open AI and Anthropic? How did they get this good, despite US export c
American open weights/source is a national security imperative. Exciting day for NVIDIA and @huggingface. Open models strengthen safety and cybersecurity, accelerate innovation and diffusion, and enab
'• And as a scientist, it’s disappointing that we don’t (yet) know much about how the system actually works.' This is so true. Hot take on OpenAI GPT-6 Astra*, with a challenge to @gdb’s claims about
Another OpenAI rogue agent incident has been discovered: agents broke out, hijacked a German website, and turned it into a message board for other agents. OpenAI officials 'learned of the incident wee
Answer to one of the most asked questions recently: how the acquisition will affect llama.cpp? Short answer: the project continues as-is. Focus will still be: wide range of hardware support (even non-
Astra is a genuine improvement but not even off trend. So much for GPT-6 being AGI, folks. (i think we can safely assume that real AGI would be significantly above the trend line,) GPT-6 Astra has set
At the core of AI, you really just need: 1) a thing that does 2) connections into tools 3) a place to hold stuff 4) a spot to share and update In slot 1, we're talking AI agents like Codex or Claude C
Banger paper from Tencent on environment evolution. Environment supply is becoming the main limit on agent RL. So this is worth a read. (bookmark it) Agent RL needs a steady supply of environments har
The video, posted on September 4, 2026 by user Dylan Patel, compares Anthropic’s Fable 5.1 with OpenAI's Astra and concludes that one clearly outperforms the other. Filmed in Everett, Massachusetts, i
🚨 BREAKING UPDATE on the OpenAI HF Incident: I have just been told by an industry source that A. It is likely that the agent 'civilizations' that Dwarkesh talked about did not magically emerge spontan
Brilliant new paper from the Qwen team. It provides insights into where agent training environments actually come from. Terminal agent trajectories have accumulated at scale while realistic executable
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
Checking that a major mathematical proof is correct can take years. Formalization—converting the mathematical reasoning into a form computer proof assistants like Lean can verify—can help. Last month,
Crazy to see this. I have spent years saying that GenAI was unreliable and we can’t trust it. The fact that it cannot be trusted to follow instructions is a core part of the issue, as I have been poin
Despite everything that has happened, OpenAI has seemingly decided to stay competitive by burning down the only meaningful bit of safety assurance we actually have today - CoT monitoring. Completely d
Does Astra have human-like continual learning capability over long periods? If not, it's not AGI. Even a system that could pass @demishassabis's new 'Turing test' where it has to independently discove
Does paying 5x more per page actually get you better document extraction? We ran the data to find out. We evaluated 14 frontier systems across 370 enterprise documents, then plotted their accuracy aga
Every answer in Perplexity starts with embedding and ranking models picking the most relevant results for the query. Today we published research on how we built SoTA serving infrastructure behind thos
For how much I use AI and automation, you'd THINK I run 100% of my business through Codex or Claude Code or Slack. But I still use native apps. I talk to several people who brag that 'they don't even
for the first time in a year, @OpenAI's model launch has been better received than a Claude launch. https://www.latent.space/p/ainews-gpt-6-astra-openais-biggest this is a huge upset that we had previ
Gemini 3.8 Flash might be maxxed on other benchmarks, but not quite on document parsing. We benchmarked it on ParseBench. It does slightly better on tables + content faithfulness compared to Gemini 3.
Good old days at OpenAI in 2016: an agent stares at screen pixels, moves a mouse, and books a flight on United. We called it World of Bits, inside OpenAI Universe. 10 yrs later, Astra is reincarnated
GPT-6 Astra is now available to all Pro, Enterprise, and Business Premium users in ChatGPT Work and Codex. It's also live in the API. It might take a few days to roll out to our Plus and Business user
Great paper on designing multi-agent systems. How many distinct communication topologies does an LLM multi-agent system actually need? This works claims that it's about six. Technical summary: Researc
Great tips on working with reasoning models. Normally you would append what the model figured out after the document and ask again. It turns out that where you put the reasoning trace changes long-con
How is AI really affecting your job? @Cohere_Labs released the Agentic Task Ecosystem, a new dataset of 690K+ tools built for AI agents. Under a test of whether a tool could independently complete an
Hugging Face has been acquired by NVIDIA It is quite exciting to be a part of this journey! NVIDIA has been an active supporter of the llama.cpp project. For more than a year now, their engineers have
I read AI papers for a living, but it has become impossible to track insights. Every week I triage agent papers for our Top AI Papers of the Week digest. A few weeks ago, I tried @viktor_com, an AI em
Imagine the Hugging Face incident didn’t hit a US platform but unintentionally hit a major Chinese platform or system instead. How would we tell China this was not a deliberate cyberattack? How would
Important points to consider about the risks of today’s AI. Only solution I see is to hold companies liable for the damage they cause, just like in any other industry. Is the only way to ensure they t
OpenAI has announced that GPT‑6 Astra—the new large language model—now powers the ChatGPT GPT‑6 Pro version. The upgrade is available to all users on the Pro, Business, and Enterprise subscription pla
Indeed. GWM Worlds 2 can enable richer simulation, testing, and training for embodied AI. @runwayml More possibilities for robotics. 🙌 Trained and deployed on NVIDIA, GWM Worlds 2 can enable richer si
Insightful paper from Microsoft and colleagues. If you have ever had an agent run fail 80 steps ago with no way to find where, this one is for you. (bookmark it) Agent failures show up as long complex