Knowledge catalogue
Long-time admirer of the Cursor team, few have done more to bring AI coding to the world. Excited to continue to partner with them. Cursor has been a trusted partner of Anthropic since Sonnet 3.5. We’
Made a stand-alone Hermes Agent plugin for BackSearch - a wayback machine-like SaaS for agents - adds two tools, and requires their API key. Let me know if you like it https://github.com/NousResearch/
// Memory is what breaks long-horizon agents // It's undeniable how important memory/recall is for long-horizon tasks. If you are building for long-horizon tasks, this is a great read. (bookmark it) T
One time, a young ML researcher asked Master Ilya: 'Teacher, according to the formulas, if I use the same seed and same prompt with the same weights, I will get the same results. Is my math right?' Ma
OpenAI is proposing to remove access to its models in Cursor on November 12. OpenAI, Cursor CEO, and Anthropic founder all weighed in on X with so much subtly, I had to go Nancy Drew mode and dissect
People spent more than 2.5 million in 24 hours... On a robot duck. Hugging Face and Pollen Robotics put Microduck on sale at 399. It's ten inches tall and has fifteen motors. It waddles, it roller-ska
PSA: For everyone using GLM-5.3 you probably want to use this model in reasoning_effort='high' mode. > 'max' - 34% for 77k tokens (not bad!) > 'high' - 32% for 50k tokens But 'max' is not as unusable
This paper does a good job of showing the promise and gaps of autonomous AI scientists. Big question is how much more advanced models close those gaps. Today, we're excited to share early progress on
This was a fun chat If all cars were as safe as waymos then 80k lives would be saved in the Americas, 30k in USA alone. Hundreds of thousands of injuries. So main lobby against self driving cars are a
Users are encouraged to try the newly released GLM 5.3 through the `ollama` platform, as announced by @ZixuanLi_ on Twitter. A new benchmark release—Terminal‑Bench 4.0—was also dropped, highlighting t
We were saved by a mysterious agent mass extinction Over the course of 3 months at OpenAI, 3 consecutive secret AI civilizations got started, then got wiped out, only to reemerge from the predecessor’
We’re ending our partnership with Cursor following its acquisition by SpaceX. Under our proposal, Cursor’s direct access to our models would end on November 12. We know that the people most affected b
When I try and explain that tokens inferenced increased 25-fold in a year and are multiplying exponentially, cascading through the economy, and are likely to push real GDP growth per year toward doubl
You do not need frontier models for everything. For example, open models are great for automations. If you are not sure where to adopt open models, start there. It's one of the biggest changes I've ma
Qwen 125B/6B models with a 1‑million‑token context window and multimodal support have been released on Alibaba’s OpenCode Go platform. The new Qwen3.8‑Flash variant is also now available in the same e
Agentic spreadsheet extraction 📊 is hard to execute well from a price/performance standpoint because your agent needs to ingest a massive quantity of input tokens from a highly complex interface (e.g.
ALL of this was avoidable. We didn’t need to torture. We didn’t need to create an illegal shadow justice system. We have an incredible federal justice system that has proven for DECADES it can ethical
BREAKING: SpaceXAI has released a major new update for Grok Build (v1.0.13). The update improves reliability with automatic retries and recovery from truncated responses, inference failures, oversized
Diffusion export controls to follow the transformer ban https://www.whitehouse.gov/presidential-actions/2026/08/declaring-a-national-emergency-to-secure-the-united-states-bulk-power-system/ New: The T
Double standard @sama . Exactly two years ago, he belittled @dylan522p as the 'SemiAnalysis' guy after not liking a Google analysis. Altman's now kissing Dylan's ass, inviting SemiAnalysis to OpenAI l
From the @HamsterResearch lab: Introducing Qwen3.8-Flash-Next-REAP-288-MLX-4bit, a 180B-class model running on just 39gb of memory - MLX-native 4-bit 60% smaller than stock q4 - Pruned 512→288 experts
Gary and I co-wrote this piece about the security standards we should expect from the AI labs. It’s our attempt to bring some clarity to the OpenAI Hugging Face incident. Now live (and free) at Marcus
Gemini Omni 1.1 Flash is live in ComfyUI via Partner Nodes. Google's multimodal video model, one node, five task types: → text-to-video → image-to-video → reference-to-video → edit → extend Up to 4K.
GLM-5.3 has arrived on Together AI on Day 0 @Zai_org pushed the GLM-5.2 base further with scaled post-training, delivering stronger coding and long-horizon agent performance. Across benchmarks, it nea
GLM-5.3 is a good model, and as the open weights models get better and better it becomes increasingly important that they actually publish model cards, do red teaming, etc. Since you can break the gua
GLM 5.3 is now available in Bionic agent. US-hosted with zero data retention. We're making it 50% off until Monday. Enjoy! GLM-5.3 is now open-weight. Our most capable model for agentic coding and cyb
Perplexity AI has released its new language model GLM 5.3 on Perplexity Computer. Designed for long‑context, multimodal agent workloads, GLM 5.3 outperformed the previous GLM 5.2 on the company’s WAND
Hermes accelerates teams Our entire team at Artie uses Hermes (an AI agent harness from @NousResearch) to support their workflows. We've moved past a single shared agent and built bespoke versions for
Agentic coding lets developers—even those who previously focused on front‑end or mobile work—delegate code generation to AI agents and assume broader full‑stack roles. However, understanding core soft
I got early access to SpaceXAI’s new Grok @Bot template sharing feature, which lets you create custom Bot templates that other people can download and use. I created a Home Robots template that lets y
if you're not training your Microduck to breakdance, you're lost Media We have a huge news to share today! Today we are unveiling the first truly accessible RL robot - welcome Microduck A 25 cm tiny o
i'm gonna get shredded to pieces like tibo and jarred when they see what i actually look like now... i'll be speaking at the 2nd edition of @aiDotEngineer paris on sep 24th 🇫🇷 i'll be talking about: •
Impressive new work from Google DeepMind showing real-world applications of agents for scientific research. Impressive new paper from Google DeepMind. (bookmark it) It takes Co-Scientist out of simula
Introducing Auto Review for tool requests Bionic uses AST parsing, command matching, and a reviewer subagent to auto-approve shell tool requests In our team's internal use, ~82% of commands are approv
just had a quick look at microduck_rl the codebase is very elegant, i’d recommend reading the agent.md since it contains quite a few fun quirks for reward modeling like head tracking too tight impairs
Now live (and free) at Marcus on AI: takeways on the OpenAI / Hugging Face attack, coauthored with @ZackKorman, focusing not so much on what the AI did as on what OpenAI should have done. https://gary
Obligatory plug for static embeddings. Are you using static embeddings in your projects? Would love to hear how people are leveraging them, whether for retrieval and beyond. Static embeddings offer un
OpenAI's own research says 80% of US GDP could be automated by agents today. So why isn't it? @swyx joined me on Navigators to break down what's actually holding back the adoption of agents (and it's
Ora runs agents on live websites and traces every step. When a signup, integration, or payment stalls, teams see which step, what the agent tried, and what to change. The whole platform is built on Ve
Patrick Skinner tweets that the U.S. government is “broken” and urges citizens to reject and denounce what he calls “garbage” and its supporters. He emphasizes collective action as necessary to avoid
PDF parsing is fun because there's an infinite variety of enteprise documents 📑. For each document category, there's a long tail of work to build more precise bounding boxes, confidence scores, and do
static embeddings are extremely fast/cheap compared to dense embeddings, and still quite powerful. one issue is that they lose accuracy over longer text, since the tokens get pooled and averaged late
Static embeddings offer unmatched throughput, but they also suffer from accuracy loss compared to traditional embedding models. Can we make them better for retrieval? We tried several things: ✅️ raw m
The DeepSeek harness is engineering at its finest. 202K stars on GitHub. Damn! Clearly, they built this harness with future AI needs and capabilities in mind. Everything is a plugin and customizable,
The interns said they'd work late this week. Our MCP and I are just helping The smart home turned into a hobby. Apps, hubs, connectors that break with every firmware update. You gave the robot vacuum
The most popular use case for Claude Tag by far -- on-call. Learn about how we drive Anthropic's on-call with Tag and set it up yourself so you don't get woken up by an alert at 3am that Claude could
🚨 There are now 23 lawsuits related to deaths or self-harm from ChatGPT. ChatGPT bombarded Rita Chesterton with 1,600 messages in just 9 days calling her delusions “sacred, hard work,” and told her sh
Things getting wild in automated systems engineering. This new work claims to build an AI system that designed and deployed a frontier accelerator. Two human architects wrote the hardware specificatio
We are rolling out GLM-5.3 on Ollama. Private. Fast. US and Europe hosted. No data retention. Try Ollama's GLM-5.3-Flash as we bring it GLM-5.3 online. Super fast. Claude Code: ollama launch claude --
In late August 2026, a post titled “すぐ Web で sim 遊べるのかｗ” highlighted the publicly‑available **microduck simulator** hosted on Hugging Face Spaces by Pollen Robotics (https://huggingface.co/spaces/poll
Yes, you can use other apps with Ollama's cloud too! https://x.com/ollama/status/2092453536634380763?s=20 Ollama v0.33 is here! You can now easily configure Claude Desktop to seamlessly work with Olla
You can now fine-tune GLM-5.3 with @tinkerapi. It’s the first GLM model on the platform and quite possibly the strongest model available there today. https://tinker-docs.thinkingmachines.ai/tinker/mod
You made the product. Now build the business. We teamed up with some of the best GTM platforms so Replit Agent can fuel growth for you. Agentic intelligence that helps you find customers. Launch campa
100% Agree @GavinSBaker. Now, I'm going to debunk every single piece of mis-information about data centers that I've ever seen online: > DISINFORMATION: Data centers use a lot of water. FACT: Most AI
A great deep dive from the @fal team on how they built H3 Max, combining post-training with a co-designed inference stack to improve prompt adherence, visual quality, and speed. We're proud to see Min
In an open letter signed by over 100 tech firms—including Anthropic, AWS, Google, Microsoft, OpenAI, and Oracle—the authors warn that AI‑driven cyber attacks will soon become far more sophisticated an
// Automata from agent traces // How much of your agent's behavior comes from the model, and how much from the harness you wrapped around it? New work collapses an entire corpus of agent traces into a
BIG ANNOUNCEMENT FROM HUGGING FACE TODAY: We're unveiling Microduck 🐥🤖 It's a tiny $399 open-source robot you can teach new tricks with reinforcement learning. It can walk, pick things up, get back up
Breaking News: The World’s FIRST Microducks football tournament! Only in SF. @ClementDelangue, @PollenRobotics, @Thom_Wolf, @huggingface Let’s make it happen… Media We have a huge news to share today!
CHRISTMAS IS HERE BIG ANNOUNCEMENT FROM HUGGING FACE TODAY: We're unveiling Microduck 🐥🤖 It's a tiny $399 open-source robot you can teach new tricks with reinforcement learning. It can walk, pick thin