10 minutes Metal Arm Station deployment setup Collab with Caldera with @yeokch @raghav2956 @HuanbinChe0dde. Ready for zero-shot rollouts with Dream-1 @dreamlabs_ai

tired of packing up your tools after using them? just get @dreamlabs_ai dream-1 and the @makermodsai metal arm to do it

Try Dream-1 on SO-101 on playground.makermods.ai

Try your SO-101s with playground.makermods.ai

The fastest way to try Dream-1 on your own robot arms: playground.makermods.ai Plug your arms into a laptop, open Chrome, type a task. Nothing to install: no Python, no ROS, no drivers No arms yet? Buy one from makermods.ai How it works 👇 The whole robot stack runs in a browser tab. Servo drivers, calibration, cameras, the control loop and IK are all TypeScript. The tab talks to your arms over Web Serial. Dream-1 runs in the cloud, everything that touches the robot runs in your browser. We wrote the SO-101's Feetech STS3215 servo protocol from scratch in TypeScript, packet by packet at 1 Mbaud. Two background threads split the work. One owns the serial ports and runs a 20 Hz control loop. The other handles the cameras, the Dream-1 calls and the IK. Setup takes a few clicks: - Swing to find: move each arm by hand and the console watches every USB port to see which one moved. No guessing /dev/tty names. - Auto-calibrate: motors drive each joint to its end stops, write the result to the servos' memory, and line it up with Dream's reference arm. Both arms calibrate at once. Dream-1 outputs hand poses, not joint angles, so something has to turn them into joint commands. We ported Dream's own trajectory IK solver to TypeScript. It solves each full 48-step action chunk at once, with velocity, acceleration and jerk limits, and it's checked against the Python SDK's output. No arm yet? The same page runs a MuJoCo physics sim in the browser, compiled to WebAssembly. Prompt Dream-1 on virtual arms first. Works with SO-101, Maker Arm and Metal Arm. try it out yourself -> playground.makermods.ai

Try @dreamlabs_ai's latest model, dream-1, live on @makermodsai metal arms at the @encord_team Physical AI Summit 🦾

This made me question: do we need fine-tuning at all? It improves performance for 10 of our 12 beta tester robots. Fine-tuning also of course helps to specialize Dream-1 to specific tasks. However, onboarding a new robot often just doesn't require fine-tuning anymore.

Day 20/365 of robot learning. I spent today experimenting with @dreamlabs_ai Dream-1 model on my so101 arm, trying to get it to complete a simple pick-and-place task without any fine tuning. Since I'm running a single arm on my macbook instead of the typical two-arm setup the model was designed around, I had to configure everything manually. After 20+ runs, a few changes somewhat improved its performance: -Experimented with mapping my so101 as the model's left arm instead of the right, which lead to fewer hallucinations. -Checked the gripper's tcp with dreamcli tcp; it needs to sit right at the fingertips (moving it inward into the gripper made things worse). -Repositioned my top and wrist cameras to better match the YAM setup the model was trained on. -Added table collision constraints and joint limits to prevent the arm from pressing into the table. In my best run so far, the arm reached my deodorant, but the gripper closed at the tip of it. This was the same issue I encountered in sim funnily enough. It worked there, but not in real life. It still hasn't completed the full pick-and-place task, but it's definitely making some progress! Also, huge thanks to @NamraByTheSea from Dream Labs for reaching out and helping me debug the setup! @dreamlabs hoping I can get some extra credits since the API wasn't working with my single arm setup :) Going to get back to reading papers soon and try out @nvidia 's newly open-sourced ASPIRE framework to experiment with agentic robot control!

Dream-1 with the Maker and Metal Arm This is the best combo to try out zero shot tasks!

Working with robots has always been harder than it should be. The intelligence was hard to access. The hardware was hard to afford. Today, both change. @makermodsai Maker Arm + Metal Arm, powered by @dreamlabs_ai 's Dream-1. Give your robot a task in plain language, right from your browser. Try it: playground.makermods.ai

Today, Dream Labs is out of stealth. In February 2026, while we were at NVIDIA, @scott_e_reed @youliangtan @GeYunhao and I started calling a new kind of robot model a World Action Model, one that learns how the world moves from video, then uses that to decide how to act. Today, we’re releasing Dream-1 through a public API that anyone in the world can use. When you rent a car you’ve never driven, you don’t learn to drive all over again. Within a few minutes, it feels like your own. Robots should work the same way: a good model should be able to pick up a new robot and get to work, without being retrained from scratch. That’s what we’ve spent the last 6 months on, and it’s what Dream-1 does. It controls bimanual robots with parallel grippers it has never seen before, zero-shot, on unseen tasks in unseen environments. Generalization has been my North Star in robotics: first by distilling video world models into VLAs (DreamGen), then by learning to act from those world models directly (DreamZero), and now by transferring across robots (Dream-1). I’ve come to believe it’s the only direction that lasts. Pushing a narrow model from 90% to 99.9% on a single task is real work, but much of it gets washed away once a model that generalizes arrives. Since the start of the company we’ve stayed single-minded about generalization.That focus is what got Dream-1 to control robots it was never trained on, on tasks it was never shown. And it’s what makes a public API possible at all: a model that only does one task on one robot can’t serve millions of users. Dream-1 is still a research beta, and it will make mistakes. But it already drives cars it’s never seen. Generalization wins in the end, and the end is closer than it looks. Here’s Dream-1 taking the wheel of a robot it’s never seen. Full details 👇

Our Metal Arm spotted 👀

Get your full teleop station for under $3K

It's crazy that a whole teleop setup, 2 leader arms + 2 follower arms, comes in under $3K. And it still lifts 1.5 kg with the same precision. Check it out → makermods.ai/maker-arm-stati…

Metal Arm is available now Get yours at makermods.ai/metal-arm 6 DoF. 3kg Payload. Full CNC construction. Ships in 1 week. DMs open.

Check out the new makermods.ai website

“Pick up the bottle and put it in the tray.” Prompt straight to action A first look at what’s coming next for Maker Arm, Metal Arm, and MakerMods Playground. Text it. Talk to it. Teach it. More soon. 👀 makermods.ai

Excited to announce I’ve joined @eptmpa to build the future ot multi-material resin 3D printing. We’ve been heads down for the last two months building our first product. We re-invented the full stack, everything from custom 3D printers to our own materials. More to come on that soon 👀👀