Zero-Shot to New Robots: Introducing Dream-1

For decades, robotics has been stuck in a chicken-and-egg problem: a robot needs to be useful before people want to buy it, but making it useful requires data from that exact robot, and that data only exists once enough people start using it. Dream-1 is our first step toward breaking that loop.

Self-driving found a way around this cold-start problem. Cars were useful from day one with zero autonomy, so millions of people drove them every day and generated the data needed to train increasingly capable models. Most other robots never had that advantage.

Two things are changing that.

First, robot hardware is getting cheap enough to reach far more people. A capable bimanual arm setup can now cost as little as $2k1.

Second, a robot may no longer need a large dataset of its own before it can be useful. If a model can generalize zero-shot to unseen environments, tasks, and even robots, a new robot can start with useful capabilities.

Put those together and the flywheel can finally start. Cheaper hardware puts more robots into the world. Generalizable models make those robots useful earlier. Useful robots generate more data, and that data makes the models better for the next robot.

02

The Cambrian explosion of cheap, affordable robots

Dream Labs enabled robots.

Robot hardware is getting cheap faster than anyone expected. See the robots above, where 7 semi-humanoid robots cost less than $15k, a price tier that barely existed a year ago. At the far end of the price curve, Hugging Face's $399 Microduck sold more than 10,000 units within 4 days of its launch in late August2. We believe the robotics take-off will start with robots like these.

$2,000

a bimanual arm setup

A capable two-arm setup, as of today.

7

semi-humanoids under $15k

Wheels, two arms, parallel grippers; a tier that barely existed a year ago.

$399

Hugging Face Microduck

Launched late August.

10,000+

Microducks in 4 days

Sold within four days of launch.2

That's because their buyers are consumers, prosumers, and small teams, not enterprises. An enterprise buyer is making a spreadsheet decision. If the robot doesn't beat a person on cost, speed, or reliability, it doesn't get bought; that's a high bar to clear. Consumers are different. 10,000+ people just paid $399 within 4 days for a robot whose main job is to be delightful and cute: it walks, waddles, and roller-skates. Somewhere between that and a fully capable home robot is a sweet spot, a robot affordable enough to be an easy yes and useful enough to earn its place in the home. Once robots hit that point, we think adoption starts to compound, and every robot in a home becomes a source of new training data. Imagine two Microducks folding your laundry, using their beaks as grippers.

03

The result that started Dream Labs

In February 2026, our team released DreamZero, the paper that introduced World Action Models3. Since then, the community has adopted WAMs as one of the most promising paradigms in robot learning. At that time, the most surprising result for us was when after post-training DreamZero on just 30 minutes of YAM robot data, we saw it generalize zero-shot to tasks it had never seen. That kind of generalization used to take hundreds or even thousands of hours of embodiment-specific data. If 30 minutes of data from a given robot was enough to transfer general-purpose task knowledge to that robot, breaking the chicken-and-egg loop no longer looked impossible. Combined with the hardware wave above, the timing felt right. That's when we decided to start Dream Labs.

After six months of building, we're excited to share a new research breakthrough that was only made possible through world action models: early signs of zero-shot embodiment generalization.

04

Introducing Dream-1

Dream-1 is our first World Action Model available through a public API that can be used all around the world. Out of the box, it can control unseen robots for unseen tasks in unseen environments.

Anvil RoboticsCaribbean

Sort the candy into containers 1 of 2

BitRobotShenzhen, China

Pull out a tissue 1 of 4

CortexSingapore

Stack the white cup on top of the pink cup 1 of 2

EastworldKL, Malaysia

Place the apple in the brown box 1 rollout

Hugging FaceParis, France

Pick up the red cube and put it in the green container 1 rollout

KAISTSeoul, Korea

Fold the box 1 rollout

XDOFJakarta, Indonesia

Find the marker and pick it up 1 rollout

RoboCurveSan Francisco, USA

Stand all three bottles upright 1 of 4

Scale AISan Francisco, USA

Stack the cubes into a single tower 1 of 9

Robo Inc.Los Angeles, USA

Move chips to tray 1 rollout

MeckaNew York, USA

Lift a test tube from the loose rack 1 rollout

AgiBotShanghai, China

Pick apple to plate 1 rollout

1–13 of 13

Autonomous · 4×

Fold the dress.

Clear the dishes from the countertop into the sink.

Pull the cap off of the blue highlighter.

Hang the mug on the mug hook.

Grab the cat litter scoop and scoop inside the litter box.

Restock the food cans from the box onto the shelf.

Put the egg in the empty spot in the egg carton.

Fold the shirt.

Put the straw into the bottle neck.

Wipe the countertop with the blue sponge.

Put the tools back in their slots.

Pour the fruits from the bowl on the left to the bowl on the right.

Slide the middle book out from the stack of three books.

Fold the green placemat in half.

Dry the plate with the towel and place it on the drying rack.

Open the box, take out the pink jump rope and close the lid.

Use the knife to push the chopped carrot to the right side of the cutting board.

Fold the long sleeve shirt.

1–9 of 18

Unseen Tasks Dream-1 was able to do. Autonomous. 8x speed at the beta testers’ sites, 4x in-house.

04.1

Get started in minutes

We want generalizable robotics to be accessible to anyone with a robot, not just large labs. So we designed Dream-1 to be as close to plug-and-play as possible. Connect your robot, point it at the API, and run. No machine learning background is needed to get started. Every week, we run internal hackathons to keep time-to-deploy as short as possible. Getting started tutorial

import dream

client = dream.DreamClient(
    api_key=<API_KEY>,
)
client.policy({"checkpoint": "dreamlabs/dream-1"})
client.set_observation(observation)
action = client.get_action()
client.close()

Condensed client side code

04.2

Zero-shot on robots it has never seen

Dream-1 can zero-shot generalize to many bimanual, parallel-gripper robots. Here are some we've tested. It can even control the original PR2, a general-purpose bimanual research robot released 18 years ago. Note that no embodiment-specific data was used to get these rollouts.

AI WorkerROBOTIS

Turn the large dial so its pointer points at the red mark

ALOHATrossen (Interbotix)

Press the big red mushroom button

ARX X5ARX

Tip the cereal box onto its side

BaxterRethink Robotics

Press the big red mushroom button

Bracket BotBracket Bot

Tip the plastic bottle onto its side

Franka Research 3, two armsFranka Robotics

Tip the cereal box onto its side

Galaxea A1ZGalaxea

Push the tennis ball into the basket

Gen3, two armsKinova

Place the umbrella into the bucket

LBR iiwa 14KUKA

Pick up the tape and place it in the round white bin

Maker ArmMakerMods

Close the glasses case

MantisBlue Worm

Pull the cable out of the box handle

Metal ArmMakerMods

Put the tape roll over the bottle

OpenArm 2.0Enactic

Close the glasses case

OpenYAMAnvil Robotics

Pick up the knife by its handle, never the blade, and place it in the tray

PiPER, two armsAgileX Robotics

Lift the pot by its two handles, carry it to the red mat and set it down level

PR2Willow Garage

Press the big red mushroom button

reBot B601-DMSeeed Studio

Pick up the knife by its handle, never the blade, and place it in the tray

Sawyer, two armsRethink Robotics

Close the top drawer

SO-101Hugging Face

Tip the plastic bottle onto its side

Trossen WorkbenchTrossen Robotics

Pick up the knife by its handle, never the blade, and place it in the tray

Unitree R1Unitree

Put the small bowl inside the large bowl

UR5e, two armsUniversal Robots

Pick up the mug by its handle and place it on the tray

WidowX 250, two armsTrossen (Interbotix)

Press the big red mushroom button

xArm 7, two armsUFactory

Tip the plastic bottle onto its side

YAM + UMI gripperI2RT x GenRobot.AI

Tip the plastic bottle onto its side

YAM-7I2RT

Tip the plastic bottle onto its side

YuMiABB

Tip the cereal box onto its side

OpenYAMAnvil Robotics

Drape the hand towel over the rack

OpenYAMAnvil Robotics

Hide a dumpling under the towel

YAM + UMI gripperI2RT x GenRobot.AI

Pull the towel to bring the plate to the center of the table

Maker ArmMakerMods

Put the sponge in the dog bed

OpenYAMAnvil Robotics

Fold the towel

Metal ArmMakerMods

Go ahead and put the banana in the bowl

reBot B601-DMSeeed Studio

Go ahead and put the banana in the bowl

YAM + Enpire grippersI2RT x Enpire

Put the straw into the bottle neck

Metal ArmMakerMods

Put the grape into the pan

OpenYAMAnvil Robotics

Close the top drawer

reBot B601-DMSeeed Studio

Push the tennis ball into the basket

OpenYAMAnvil Robotics

Put the baby bottle in the pen holder

Maker ArmMakerMods

Pull the towel to bring the plate to the center of the table

OpenYAMAnvil Robotics

Grab a handful of cotton balls and put them in the cup

OpenYAMAnvil Robotics

Set the tennis ball on top of the upside-down mug

Metal ArmMakerMods

Stack the bowl inside the small pot

OpenYAMAnvil Robotics

Put the plate on top of the basket like a lid

OpenYAMAnvil Robotics

Hang the headphones on the hook

YAM + UMI gripperI2RT x GenRobot.AI

Put the fruit that is not the kiwi in the basket

reBot B601-DMSeeed Studio

Pick up the mango and place it in the bowl

Vega 1 Pro + YAM GripperDexmate x I2RT

Pick up all markers one by one and put into black pencil holder using both arms

Unseen Robots Dream-1 was able to control. 4x speed on real robots, 3x in simulation.

This is one of the first reported zero-shot generalizations to completely unseen robot embodiments.4 If you have a bimanual parallel-gripper robot, try prompting Dream-1 and let us know how it goes. Zero-shot performance depends partly on your robot's inverse kinematics (IK) solver, so we've written a guide to help you get the most out of it. IK guide. Performance also varies across robots. Below, we break down zero-shot performance in simulation and on real robots.

Dream-1 zero-shot performance on 27 unseen robots in simulation and 4 real robots, relative to YAM, the robot Dream-1 was actually trained on. The real robots ran earlier Dream-1 checkpoints that never trained on them, each compared with the same checkpoint on YAM; OpenYAM was tested on unseen tasks only. Simulated evaluations are conducted by doing 300 rollouts per embodiment: 150 rollouts of 50 tasks quantifying language following and 150 rollouts of 150 tasks quantifying unseen task generalization.

Below, we explain how zero-shot embodiment generalization works and where it falls short. We then show how one-shot embodiment learning closes that gap.

04.3

How zero-shot embodiment generalization works

Two ways to control a robot: joints or end-effectors. Robot policies commonly output joint angles, which ties them to one specific robot's body. Dream-1 can do the same for the embodiments it supports natively, but it can also predict end-effector (EEF) poses: where each gripper should be and whether it should be open or closed. Every robot with two arms and parallel grippers shares that EEF action space, and its own IK solver turns the poses into joint commands. That shared interface gives Dream-1 a common way to command robots it has never seen, but on its own it isn't enough to make zero-shot transfer work.

Why does this work? Dream-1 sees the world through three cameras: a head view and a wrist view on each arm. Prior work on the Universal Manipulation Interface (UMI) showed that policies can transfer zero-shot across robot arms when every arm uses the same gripper and the policy relies only on wrist cameras5. From the wrist's point of view, the robots look identical. But that shortcut has a hard limit. It only works when the gripper hardware matches exactly, and it gives up the head camera's view of the whole scene.

World Action Models remove that limit. A WAM learns to predict video and actions together, so controlling a robot and imagining what that robot will do are the same problem. And every robot shows up the same way: as pixels. The head camera sees a UMI gripper, a two-finger gripper or an arm Dream-1 has never met as nothing more than shapes moving through a scene, which makes video the one representation every robot type shares. If Dream-1 can predict what a robot will look like as it completes a task, it can produce the actions that get it there. That’s why it can control robots with grippers that look nothing like UMI’s, and even entirely new bodies seen through the head camera.

How a World Action Model works.

Limitations

Predicting EEF trades some control for generality. Compared with joint-angle prediction with per-embodiment adaptation:

• Performance depends on the quality of each robot's IK solver.
• Dream-1 doesn't control how the arm reaches a pose, only where the gripper ends up, which can matter around obstacles or near joint limits.
• Robots whose workspace or kinematics / dynamics differ sharply from what Dream-1 has seen may predict future states that are infeasible.

04.4

What Dream-1 can and can't do

Robocurve’s independent report on Dream-1: mean score per task attempt on RoboDojo-RC Tier 1.

Ahead of launch, Robocurve6, a Public Benefit Corporation that measures and reports frontier robotics capabilities to the public, evaluated Dream-1 independently. Their report, above, shows Dream-1 is better, faster and cheaper: a new Pareto frontier for robot control. It scores above the next best models, Opus 5.5 and Gemini 4 Argon, at 16x lower cost per task attempt than Gemini 4 Argon and 1.7x faster than the fastest LLM, GPT-6 Astra. While current LLMs excel at planning, robots still require a fast, general System 1 to act, and that is what Dream-1 is built to be.

How much does it cost to run Dream-1? We want you to know exactly what to expect before you try it. Running the model costs just $3 per hour for Dream-1, or $2 per hour for Dream-1-flash. Each call to the model returns 1.6 seconds of actions, so an hour is 2,250 calls: about $0.0013 an action for Dream-1, and $0.0009 for Dream-1-flash.

$3 / hour

Dream-1

Everything above the hourly cost is yours to keep.

$2 / hour

Dream-1-flash

The faster, lighter model.

Across tasks. We break Dream-1’s performance down into eight task categories, from what it handles well to where it still struggles.

Fine-tune for your task with FaaS. If Dream-1 falls short on a task you care about, you don’t have to wait for our next release. Collect some data on your own robot, run it through FaaS, and get a custom model built for that task.

We want anyone with a robot to work like a forward-deployed engineer: find a task someone needs done, collect a bit of data, adapt Dream-1 to it, and deploy. If people will pay for what your robot does, you keep everything above the hourly cost.

Dream-1 is a research beta

It will make mistakes, and it has not been validated for safety-critical use. Please test it in controlled settings, and take extra care whenever a robot is running near people. If you see unsafe behavior, please report it to safety@dreamlabs.ai.

04.5

One-shot embodiment learning

When zero-shot isn't enough, we fine-tune Dream-1 for your robot. We call this Fine-tuning-as-a-Service (FaaS).

Here's what that looks like in practice: timelapses of data collection on two of our partners' robots, 5 minutes each.

Data collection timelapse

Five minutes of data collection on two robots at 8x: reBot B601-DM on the left, Maker Arm on the right.

After adapting on just those 5 minutes, the difference shows side-by-side with the zero-shot run. Dream-1 now understands a bit better the affordance and kinematics of the robot.

reBot B601-DM

Simulated twin

Zero-shot

Time0:00

+ 5 min

After FaaS

Time0:00

Maker Arm

Real robot

Zero-shot

Time0:00

+ 5 min

After FaaS

Time0:00

Zero-shot and after FaaS on the same instruction, “go ahead and put the banana in the bowl”, with five minutes of the robot’s own data in between: reBot B601-DM in its simulated twin, Maker Arm on the real robot. 3x speed.

Some embodiments sit much further from what Dream-1 already knows, and closing a gap that large takes more of the robot’s own data. Dexmate with Sharpa dexterous hands grasps with hands in place of grippers, and took 77 minutes of its data. On YAM, we swapped in different end effectors for this task, a compliant gripper and a modified knife gripper, and it took 98 minutes of data.

After FaaS5x speed

Dexmate, Sharpa dexterous hands, 77 minutes of data

Pick bread from toaster and place on plate

After FaaS5x speed

YAM, compliant gripper and modified knife gripper, 98 minutes of data

Cut the leaf off the strawberry

After FaaS on embodiments far from Dream-1’s training data, on the real robots. 5x speed.

Across twelve robots from our FaaS beta partners, fine-tuning lifts the language-following score on ten of them:

Zero-shot versus after FaaS on the same twelve robots, in simulation, relative to YAM.

One-shot adaptation still outperforms zero-shot, so a gap remains. Closing it is one of our main goals: the better zero-shot gets, the closer any new robot comes to working out of the box.

05

Looking forward

Dream-1 today is a manipulation policy for bimanual robots with parallel grippers. It’s a starting point. We believe a strong, general robot foundation model can extend to mobile manipulators, dexterous hands, and bodies we haven't imagined yet, opening up many more of the things people actually want robots to do. Robotics won't be solved on day one, and Dream-1 has real limits, which we've tried to show transparently above.

Robotics is still stuck in a difficult loop: every new robot or task demands new data and a custom policy, often with limited generalization. With Dream-1 as a robotics API, we want to break that loop and put robot intelligence into more people's hands, so everyday users, not only researchers, can discover new applications for general-purpose robot intelligence.

Dream-1 . public API

Put Dream‑1 on your robot

If you're building robots or putting them to work on real tasks, we want Dream-1 to be the fastest way to make them useful. Tell us what you're working on.

Use Dream-1 Today

Dream-1 MicroDuck control

A fine-tuned Dream-1 checkpoint controlling multiple embodiments, including the Microduck.

Notes and references

1. 1MakerMods, Maker Arm: a bimanual arm setup from about $2k. ↩
2. 2Yahoo Tech, Hugging Face's $399 Microduck robot has already sold more than 10,000 units, 1 September 2026: more than 10,000 units since pre-orders opened on 28 August. ↩
3. 3DreamZero: the paper that introduced World Action Models, February 2026. ↩
4. 4LAP also showed signs of life of embodiment generalization: Zha et al., LAP: Language-Action Pre-Training Enables Zero-shot Cross-Embodiment Transfer. ↩
5. 5Chi et al., Universal Manipulation Interface: In-The-Wild Robot Teaching Without In-The-Wild Robots, RSS 2024: zero-shot transfer across arms that share one gripper, from wrist cameras alone. ↩
6. 6Robocurve, a Public Benefit Corporation measuring and reporting frontier robotics capabilities to the public: Dream-1 on RoboDojo-RC Tier 1, October 18, 2026, CC BY 4.0. ↩

Cite this post

@misc{dreamlabs2026dream1,
  title  = {Zero-Shot to New Robots: Introducing Dream-1},
  author = {Dream Labs},
  year   = {2026},
  url    = {https://dreamlabs.ai/research/dream-1}
}