给Dream-1一个带有平行抓手的新双人机器人，将其连接到我们的API上，它可以开始在盒子外执行任务。 没有微调，没有从机器人的演示，和不需要机器学习的专门知识。
这是走向通用机器人智能的重要一步。
今天，梦幻实验室也以90000M的种子资金从隐形处出来，并通过一个公开的API让梦幻-1提供给世界各地的人们。
01
解决鸡蛋问题
数十年来，机器人一直被困在鸡蛋问题中：机器人在人们想要购买之前需要有用，但让它有用需要来自那个精确的机器人的数据，只有足够多的人开始使用它时，数据才会存在。 梦想1是我们打破这个循环的第一步。
自驾车找到了解决这个冷启动问题的方法。 从零自主的第一天起，汽车就变得有用，所以每天有数百万人驾驶汽车，并生成培训能力越来越强的模型所需的数据。 多数其他机器人从未有过这种优势。
有两件事改变了这一点。
首先，机器人硬件越来越便宜，可以接触到更多的人。 一个有能力的双人臂架现在可以花费2千1美元。
第二，机器人在发挥作用之前可能不再需要自己的大数据集。 如果一个模型能够将零镜头概括到看不见的环境，任务，甚至机器人，那么一个新的机器人就可以从有用的能力开始。
把这些放在一起，飞轮终于可以开始了。 廉价硬件让更多的机器人进入世界。 通用的模型使得这些机器人更早有用。 有用的机器人生成更多的数据，这些数据使得模型更适合下一个机器人。
02
Cambrian爆炸廉价、廉价的机器人
机器人硬件越来越便宜比预期的要快。 参见上面的机器人，其中7个半人造机器人的成本低于15k，这一价格水平在一年前几乎不存在。 在价格曲线的远端，Hugging Face的399Microduck在8月2日下旬的发布的4天内售出了1万多台。 我们相信机器人起飞会从这些机器人开始。
$2,000
双臂布置
一个有能力的两臂架设，从今天开始。
7
15k以下的半人体
轮子双臂平行的抓手一年前才存在
$399
Hugging Face 微多克语Name
发布 8月下旬开机。
这是因为他们的买家是消费者、投机者和小团队，而不是企业。 企业买方正在作出电子表格决定。 如果机器人不以成本，速度，或可靠性击败一个人，它就不会被买下；那就是一个高杠来清除。 消费者不同。 1万多人在4天之内就付了399美元买一个机器人，其主要工作是讨人喜欢和可爱：走走走走走走走走走走走走走走走走走走在那个和完全有能力的家用机器人之间的某个地方，是一个甜点，一个能够负担得起，可以轻松地是的机器人，并且能够用得上在家中获得自己的位置。 一旦机器人到达这个点，我们认为收养开始复合，家里的每个机器人都成为新的训练数据的来源。 想象一下，两个微调师在折叠你的衣服，用他们的喙作为抓手。
03
启动梦幻实验室的结果
2026年2月，我们的球队发布 DreamZero，即推出世界行动模式3的论文。 此后，社区将WAMs作为机器人学习中最有希望的范式之一。 那时，我们最令人惊讶的结果是，仅仅用30分钟的YAM机器人数据在训练后DreamZero之后，我们看到它将零镜头概括为从未见过的任务。 这种概括过去需要数百甚至数千小时的具体化数据。 如果给定机器人的30分钟数据足以将通用任务知识传递给该机器人，那么打破鸡蛋圈就不再看起来不可能了。 加上上面的硬件波，时间感觉是正确的。 那时我们决定开始做梦实验室。
经过半年的建设，我们兴奋地分享了一个新的研究突破，这个突破只是通过世界行动模式才得以实现的：零弹孔化的早期迹象。
04
引入梦想一
Dream-1是通过一个公众API提供的首个世界行动模式，可以在世界各地使用。 从盒子里，它可以控制看不见的机器人在看不见的环境中进行看不见的任务。
04.1
几分钟后开始
我们希望所有拥有机器人的人都能使用通用机器人，而不仅仅是大型实验室。 因此我们设计了Dream-1，尽可能接近插座和游戏。 连接你的机器人，指向API，然后运行。 启动不需要机器学习背景。 每星期我们都会运行内部黑客尽可能缩短部署时间开始教学
导入梦
客户端=梦想。 梦想客户端( DreamClient)
api_key=<API_KEY>,
)
客户端。 policy({"检查点":"梦想labs/梦想1")}).
客户端。 set 观察(观察)
动作=客户端.get action()
客户端。 close ()
压缩客户端代码
04.2
对机器人零镜头，它从未见过
梦-1可以将零射一般化为许多双人，平行的gripper机器人。 这里有一些我们已经测试过。 它甚至可以控制18年前的原始PR2，即通用双人研究机器人发布。 请注意，没有使用具体具体体现的数据来推广。
这是首次报道的零射一般化以完全看不见的机器人化身之一。 如果你有一个双手平行的机器人，尝试激发梦1，让我们知道它是如何进行的。 零射的性能部分取决于你的机器人反动(IK)解析器，所以我们写了一个指南来帮助你从中获得最大的解析。 IK导。 各种机器人的性能也各不相同。 下面，我们在模拟和真正的机器人上破解零射的性能。
下面，我们解释零弹射一般化是如何工作的，它在哪里不足。 然后，我们展示一枪一弹的学习如何弥合这一差距。
04.3
零弹孔化一般化是如何工作的
控制机器人的两种方法：关节或终端效应器。 机器人政策通常输出关节角，它们与一个特定的机器人身体联系在一起。 Dream-1可以对其本土支持的浮雕做同样的事情，但是它也可以预测端效应器(EEF)的外形：每个抓手应该在哪里以及是否应该打开或关闭。 每个拥有两只手臂和平行抓手的机器人都共享EEF动作空间，其自己的IK解析器将姿势转化为联合指令。 这种共享界面让Dream-1拥有了它从未见过的指挥机器人的共同方式，但单靠自己还不足以使零镜头传输工作成功。
为什么这样有效？梦-1透过三台相机观察世界：头视和手腕视每只手臂。 此前关于通用操纵界面(UMI)的研究表明，当每只手臂都使用相同的抓手时，政策可以将零射弹传递到机器人的手臂上，而政策只依靠腕相机5. 从手腕的角度来看，机器人看起来完全一样。 但捷径有困难的极限只有在握手硬件精确匹配时才有效，它放弃了头部镜头对整个场景的视线。
世界行动模式取消了这一限制。 一个WAM学会一起预测视频和动作，所以控制一个机器人，想象那个机器人会做什么，也是同样的问题。 而每个机器人显示都以同样的方式上升：和像素一样。 头部摄像头看到一个UMI的抓手，一个双手指的抓手或者一个手臂的Dream-1从来没有像形状一样在一幕中移动过，这使得视频成为了每一个机器人类型的股份的其中一种。 如果Dream-1可以预测一个机器人在完成一个任务时会是什么样子，它可以产生得到它的行动。 所以它可以控制机器人
04.4
做梦一号能做也不能做
领先于发布，Robocurve6，一家测量和报告前沿机器人能力的公益公司，独立评价Dream-1. 他们的报告，上面，显示 Dream-1更好，更快，更便宜：一个新的用于机器人控制的Pareto前沿。 它的得分高于次佳型号，即Opus 5.5和Gemini 4 Argon，每任务尝试16x成本低于Gemini 4 Argon，比最快的LLM,GPT-6 Astra快1.7x. 虽然目前的LLMs在规划上表现优异，但机器人仍然需要一个快速，通用的系统1来行动，这就是梦1所构建的。
做梦1要多少钱？我们想让你知道在你尝试之前该期待什么运行的模型只花费每小时3美元做梦1，或者每小时2美元做梦1-flash. 每个给模型的呼叫返回1.6秒的动作，所以一个小时是2,250个呼叫：大约0.0013美元是梦1的动作，0.0009美元是梦1-闪电。
3美元/小时
梦想页：1
超过小时成本的一切都归你了
2美元/小时
梦幻一闪
越快越轻
穿越任务我们把Dream-1的表现分成八个任务类别
微调为您的任务与FaaS。 如果梦一号没有完成你关心的任务，收集一些关于你自己的机器人的数据，通过FaaS运行，并获得一个定制模型来完成这项任务。
我们希望任何拥有机器人的人都能像前方部署的工程师一样工作：找到一个人需要完成的任务，收集一些数据，适应梦想1，并部署。 如果人们会为你机器人的工作付出代价，你将一切维持在小时成本之上。
04.5
一弹一弹的表演学习
当零镜头不够时，我们微调 Dream-1为你的机器人。 我们称之为微调-as-a-Service(FaaS).
在实践中是这样的：我们两个伙伴的机器人的数据收集的时滞，每人5分钟。
5分钟后，显示与零射线跑的比对。 梦1现在更了解机器人的承受力和运动力。
一些浮雕与梦1已经知道的相距甚远，Dexmate与夏帕的柔软手握着手代替抓手，并用77分钟的时间获取其数据。 在YAM上，我们交换了不同的终端效应器来完成这个任务，一个符合要求的握手和一个经过修改的刀柄握手，这需要98分钟的数据。
微调从我们的FaaS β伙伴中举出了其中10个语言跟随得分：
一发的适应性仍然超过零发，所以差距仍然存在。 关闭它是我们的主要目标之一：零射得越好，任何新的机器人都就越接近工作。
05
展望未来
今天的梦-1是双人机器人的操纵政策，有平行的抓手。 这是起点。 我们相信一个强大的，一般的机器人基础模型可以扩展到移动操纵者，狡猾的手，和我们还没有想象到的身体，打开更多的东西人们实际上希望机器人做。 机器人在第一天就解决不了梦1有真正的极限我们试图在上面透明地展示
机器人仍然被困在困难的循环中：每一个新的机器人或任务都需要新的数据和自定义政策，往往一般化程度有限。 随着Dream-1作为机器人API，我们希望打破这个循环，将机器人智能投入更多的人手中，因此日常用户，不仅仅是研究人员，都可以发现通用机器人智能的新应用。
梦-1. 公共API
把梦1放在你的机器人上
如果你正在建造机器人，或者让他们去完成真正的任务，我们希望梦1成为最快的方法，使其有用。 告诉我们你在做什么。
指出和参考文献
- 1MakerMods, Maker Arm：双臂布置，约2千元。
- 2Yahoo Tech,Hugging Face的399Microduck机器人已经售出10000多台，2026年9月1日：自8月28日预购开通以来超过10000台。
- 3DreamZero：推出世界行动模式的论文，2026年2月。
- 4LAP还显示浮雕的生命迹象
本文由 AI图鉴自动抓取并翻译，版权归原作者所有；如需查看原文请点下方链接。
Give Dream-1 a new bimanual robot with parallel grippers, connect it to our API, and it can start performing tasks out of the box. No fine-tuning, no demonstrations from that robot, and no machine learning expertise required.
This is an important step toward general-purpose robot intelligence.
Today, Dream Labs is also coming out of stealth with $90M in seed funding and making Dream-1 available to people all around the world through a public API.
01
Breaking the chicken-and-egg problem
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
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.
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.
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.
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.
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.
04.4
What Dream-1 can and can't do
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.
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.
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.
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.
Across twelve robots from our FaaS beta partners, fine-tuning lifts the language-following score on ten of them:
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.
Notes and references
- 1MakerMods, Maker Arm: a bimanual arm setup from about $2k. ↩
- 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. ↩
- 3DreamZero: the paper that introduced World Action Models, February 2026. ↩
- 4LAP also showed signs of life of embodime
Auto-fetched and translated by AI Atlas. All rights belong to the original author.