In this amazing multidisciplinary collaboration, we report our early experience with the @openclaw -> https://t.co/THXYyajfQB

@erusev @antoniostoilkov Still early — feedback welcome! Check out LlamaBarn at https://t.co/f5zVgNyf7X

I've started a company: https://t.co/jFknDoasSy From a fun side project just a few months ago, ggml has now become a useful library and framework for machine learning with a great open-source community

Introducing LM Link ✨ Connect to remote instances of LM Studio, securely. 🔐 End-to-end encrypted 📡 Load models locally, use them on the go 🖥️ Use local devices, LLM rigs, or cloud VMs Launching in partnership with @Tailscale Try it now: https://t.co/Vl2vr6HlF5

We’re building an LLM chip that delivers much higher throughput than any other chip while also achieving the lowest latency. We call it the MatX One. The MatX One chip is based on a splittable systolic array, which has the energy and area efficiency that large systolic arrays are famous for, while also getting high utilization on smaller matrices with flexible shapes. The chip combines the low latency of SRAM-first designs with the long-context support of HBM. These elements, plus a fresh take on numerics, deliver higher throughput on LLMs than any announced system, while simultaneously matching the latency of SRAM-first designs. Higher throughput and lower latency give you smarter and faster models for your subscription dollar. We’ve raised a $500M Series B to wrap up development and quickly scale manufacturing, with tapeout in under a year. The round was led by Jane Street, one of the most tech-savvy Wall Street firms, and Situational Awareness LP, whose founder @leopoldasch wrote the definitive memo on AGI. Participants include @sparkcapital, @danielgross and @natfriedman’s fund, @patrickc and @collision, @TriatomicCap, @HarpoonVentures, @karpathy, @dwarkesh_sp, and others. We’re also welcoming investors across the supply chain, including Marvell and Alchip. @MikeGunter_ and I started MatX because we felt that the best chip for LLMs should be designed from first principles with a deep understanding of what LLMs need and how they will evolve. We are willing to give up on small-model performance, low-volume workloads, and even ease of programming to deliver on such a chip. We’re now a 100-person team with people who think about everything from learning rate schedules, to Swing Modulo Scheduling, to guard/round/sticky bits, to blind-mated connections—all in the same building. If you’d like to help us architect, design, and deploy many generations of chips in large volume, consider joining us.

@shaoruu @cursor_ai This is great - I just added a council feature to our internal chat platform and next up was exploring how to use it for coding use cases!

As Rocks May Think: an interactive essay on thinking models, automated research, and where I think they are headed. Enjoy! https://t.co/dcFYrXUQYg

@wesmckinn Looks really slick and very necessary. Do you tend to develop with one agent and then have it review with another? I’ve been thinking of building with CC and reviewing with Codex

Droids can now take on Missions: goal-oriented work that may take days to fully spec out, build, and test. Genuinely mind blowing to see what Missions some of the world's largest enterprises have been sending Droids on. My favorite examples have been: • Modernize a 40-year-old COBOL core module • Migrate >1k microservices to a new Kubernetes cluster across three regions • Recalculate 10 years of pricing data after a revenue recognition rule change • Refactor a monolith that processes 20M daily API calls, with no downtime Try sending a Droid out on a Mission today and let me know what you think!

Comparing the AI open source ecosystem in the US, China, Europe, and the rest of Asia over time https://t.co/XFUgGoYKlY

We rebuilt Next.js in a week. No, really. The team ported the framework to run natively on Workers to prove what’s possible with edge-first architecture. Dive into the technical hurdles we solved to eliminate Node.js dependencies. https://t.co/GqYBiZ5Qum

Crazy project turns AI history into structured data and publishes it to Hugging Face. "DataClaw parses session logs, redacts secrets and PII, and uploads the result as a ready-to-use dataset." https://t.co/WECC3QnRsk

PyTorch Day India 2026 session recordings are now available. On February 7, 460 builders gathered in Bengaluru for a full day of technical talks focused on kernels, compilers, inference, and production-grade AI systems. Co-hosted by IBM, NVIDIA, Red Hat, and PyTorch Foundation, the event highlighted heterogeneous compute, platform stability, and end-to-end performance from edge to data center. 🔗 Full event playlist: https://t.co/0ELH0O0Ut3 #PyTorch #AIInfrastructure #OpenSource

New to the PyTorch Ecosystem Landscape: Kubetorch. Kubetorch enables ML research and development on Kubernetes across training, inference, RL, evals, data processing, and more, in a simple and unopinionated package. Learn more: https://t.co/YadOKc3sQo #PyTorch #Kubernetes #MLOps #AIInfrastructure

Mercury 2 is live 🚀🚀 The world’s first reasoning diffusion LLM, delivering 5x faster performance than leading speed-optimized LLMs. Watching the team turn years of research into a real product never gets old, and I’m incredibly proud of what we’ve built. We’re just getting started on what diffusion can do for language.

Introducing Radar, real-time collective intelligence for web agents 🧵 In our tests, agents that get a cache hit on Radar avoid the need for a browser entirely, dramatically improving agent performance Built & tested with @convex, @daytonaio and @superset_sh, @browser_use https://t.co/bFaFj6aMAA

Claude renovated my GitHub homepage for me by automatically setting up a CRON that pulls in my latest blog posts, and found images and other details to make things a bit nicer :) https://t.co/42GrcQ4w05

Claude Code now supports auto-memory. This is huge!

@garrytan This is why I love using skills. You can dynamically and more efficiently give your coding agents the data it needs. Put triggers in between to capture essential lessons between sessions and incorporate them right back. Coding agents don't have to starve anymore.

@cwolferesearch @_theopompus Does it behave closer to SFT or DPO in the k=1 case?

@LucaViano4 Does this mean it requires a custom inference stack to deploy since it’s a custom architecture with the ensemble of LoRA adapters?

This feels like NEFTune for RL.

Meta presents VecGlypher Unified Vector Glyph Generation with Language Models paper: https://t.co/anAFlgLMMV https://t.co/Nh3OpUBwa9

I’m giving an agent control over Reachy Mini from @huggingface and letting it understand and share spatial data via @Spectacles AR is the human interface for robotics and physical AI imo. It feels like absolute magic to interact with this, both in voice/agent and “puppeteering” mode. I’ll probably work on AR for either an arm (manipulation tasks) or some sort of drone (locomotion in 3D space) next… Project is fully open source btw: https://t.co/pmkXJR0U7f Thank you @SensAIHackademy for sending me the robot!