All entries

GLM-5.2 has been the most popular new model on Fireworks this past week. @ArtificialAnlys confirms why: #3 overall on GDPval-AA (1524 Elo), #1 open weights by 116 points. Interest is showing no signs

GLM-5.2 leads open weights models and sits at #3 overall on GDPval-AA, a real-world agentic work benchmark GLM-5.2 from @Zai_org scores 1524 Elo on GDPval-AA, which measures performance on real-world,

GLM 5.2 x AWS GLM-5.2 is accessible via http://Z.ai GLM API on AWS Marketplace 🚀 Powerful long-horizon autonomous workflows, top-tier coding & multi-step agent reasoning capability, delivered through

GLM is the kind of model that revives serious interest in open source AI. It passes the blind test relative to the frontier models on the median production grade knowledge worker task. It’s affordable

Hugging Face was approaching milestones of 3 million public models and 1 million public datasets on its platform, reflecting rapid growth in open-source AI resources. The announcement highlights the e

GPU rental prices have declined according to a GPU rental index tracked by Gary Marcus, indicating decreased demand or increased supply in the cloud computing market. This trend suggests potential shi

Gray Swan: Red-Teaming after Mythos & the coming AI security crisis https://www.latent.space/p/gray-swan @GraySwanAI cofounders @zicokolter and Matt Fredrikson explain why AI security is fundamentally

Great report on LLM agent communication protocols. Communication is a huge bottleneck in multi-agent systems. (worth bookmarking) The report builds a five-dimensional taxonomy (counterparty, payload,

Grok Build upgrades /goal is live on Grok Build. We use a team of agents: - implementors - skeptics - code reviewers - planners and a mix of grok build and composer in various roles. Would love to hea

Guess which is Fugu Ultra? This is how recent models compare when generating endless procedural terrain (using Three.js). All of these are one-shotted! Just wild! Trying a few more examples. Will shar

have been thinking a bunch about model routing and related things current thoughts here, would love feedback: 1/ there is a difference between 'model routing' and 'model council' 'model routing' = rou

Here is the prompt method behind this AR try-on app. The trick is not a magic prompt. It is the architecture of the prompt, and it works across GLM-5.2 and other frontier models. Full prompt: http://c

Nous Research announced that their Hermes Agent project has achieved 200,000 GitHub stars, acknowledging the contributions of developers, supporters, and users who helped reach this milestone. The ann

Hermes Agent, developed by Nous Research, has expanded its computer use capabilities to support Windows and Linux operating systems in addition to its existing macOS functionality. This enhancement, i

How does it work? Sakana Fugu is itself an LLM, trained to call various LLMs in an agent pool, including instances of itself recursively. Fugu dynamically orchestrates the world's best models to tackl

Gary Marcus discusses potential competitive threats to CoreWeave, a GPU cloud computing company, from SpaceX's large GPU contracts and infrastructure investments. The post likely examines how SpaceX's

Human intelligence is fundamentally a collective intelligence. We solve complex problems by participating in a vast cultural network that builds upon ideas across generations. I believe the strongest

This post appears to announce or celebrate the discovery or successful implementation of Ollama, a tool for running large language models locally. Based on the enthusiastic tone and source, it likely

I have been trying Sakana Fugu Ultra-high and, first, it is incredibly slow: my typical coding tests (shaders, interactive scenes) take 30 minutes to run And the results are... fine. It does not match

This post references a bill to prohibit stock trading by members of Congress, which the author claims to have introduced. The post criticizes the House Republican leadership for not scheduling a vote

Pamela Fox created a custom t-shirt design referencing Simon Willison's Pelican benchmark, a technical tool or metric in web development or performance testing. She posted about it on X (formerly Twit

If you are building agents, Max Agency with @hwchase17 is THE best behind the scenes of how the top agents are built at companies like @SierraPlatform @RampLabs @benchling @ListenLabs @_hex_tech @coge

This entry references a Pew Research Center report examining American public attitudes toward artificial intelligence, likely covering survey data on awareness, adoption, concerns, and demographic var

I’m excited to partner with @swyx to host a workshop at the @aiDotEngineer World’s Fair with Carole Robin (Distinguished @StanfordGSB Teacher of “Touchy Feely” & Co-Founder of @LITfellows) on developi

I’m running a 3 hour advanced workshop at AI Engineer World’s Fair! 🚀 2026 has greatly changed how one should learn lower-level technicals like kernels, agentic RL, reward hacking, cont learning. What

Introducing Sakana Fugu: A full multi-agent orchestration system accessible via a single model API. Our ‘Fugu Ultra’ model matches the performance of Fable and Mythos, delivering frontier capability w

Together AI conducted a blind test comparing two landing pages—one created by GLM 5.2 and one by Opus 4.8—to evaluate whether users could distinguish between AI-generated designs. The test highlights

Is it really worth it joining a robotics hackathon? One year ago, thousands of people around the world spent a single weekend building AI robots using the open-source LeRobot stack. Many had never bui

Is there a market where folk are predicting when Reflection AI will drop their first model This is probably as much compute as currently used by all the Chinese open source companies together (more ad

It seems almost too dumb to be true, but apparently the literal belief of SaaS bears is 'all software is a 0 because Claude can one-shot these apps' Just staggering levels of short-sightedness in that

Its been 10 days and the situation with Fable remains essentially just as confusing. (There have been many contradictory reports and articles and posts from different parties, which does not lower the

It’s much more likely cutting USAID SAVED 4.5 million kids lives by no longer funding the dozens of USAID-funded civil wars that had been raging across Africa, Central Asia, LatAm and beyond. Elon Mus

it's official, I'm on community at NousResearch. I'll be shipping tested walkthroughs, answering questions and carrying what the community builds back to the team. if you're on Hermes Agent, the Disco

@jerryjliu0 I also built a fullstack PDF parser using LiteParse, which is super fast. https://x.com/itsafiz/status/2068647234691026970 Built a super fast PDF parsing service with LiteParse! LiteParse

Join us in our Discord this time tomorrow for Office Hours! Our topic will be the @NVIDIAAI × @stripe × @NousResearch Hermes Agent Accelerated Business Hackathon which concludes at the end of the mont

Join us Tuesday, June 30th, for the inaugural Agent Open pickleball tournament! 🏟️ Our co-founder, @travers00, will join leaders from @Pragmatic_Eng, @braintrust, @llama_index, @turbopuffer, and @moda

Just a glimpse of what collective AI intelligence will bring. We haven’t truly cracked multi-agent orchestration but with every new frontier model, intelligence should compound. Introducing Sakana Fug

LLMs can’t be trusted to follow rules. Which means they can’t be trusted, period. If we are going to solve alignment we must move on. Large language models can be persuaded to break their own rules. N

I can't provide a summary for this entry as the magnet link appears to be a potentially malicious download (watering-hole.zip), which suggests a security threat rather than legitimate ComfyUI content.

Major kudos to @hardmaru and @SakanaAILabs on the launch 🚀 'our core conviction has been that the most powerful AI systems will be collaborative ecosystems, not isolated monoliths...the future belongs

Maturing is realising that absolutely nothing of substance is happening here. A false pressure valve is being released, to reset public anger so as to buy additional years of time to implement an agen

Mistral officially has 1,000 team members around the world! Thank you to our incredible talent for believing in the mission and contributing to what we are building each and every day. You too can be

Multiculturalism is sold as neutral cultures peacefully coexisting side by side. That’s deliberate propaganda. Some cultures are not neutral. They are openly hostile, teaching that outsiders, especial

My parallel agent side-project today was having Claude Code port the new Moebius image pinpointing model to ONNX in order to run it entirely in the browser https://simonwillison.net/2026/Jun/22/portin

Allie K. Miller expresses subjective observations about perceiving changes in Claude Opus 4.8's behavior or capabilities, describing an intuitive sense ('spidey feeling') that the model feels notably

Neuralink now has its 26th recipient Sgt. Lee Marten, @Canuckula, is a Vancouver Police Sergeant, cyclist, and ALS patient who became the first Canadian to receive the N1 brain-computer interface Lee