Xiaomi-Robotics-1

Breaking the data barrier. Scaling robot policy models with embodiment-free pre-training. 打破数据壁垒。通过无具身预训练扩展机器人策略模型。

Foundation models in language and vision keep moving the frontier by riding empirical scaling laws: capability tracks data, parameters, and compute. Robotics has missed out. Large-scale, high-quality data is hard to come by, and that scarcity, more than anything else, has capped how far policy models could scale. What robots can do under genuinely large-scale training remained largely an open question. We take a step toward answering it. Xiaomi-Robotics-1 combines large-scale embodiment-free (UMI) pre-training with a modest amount of real-robot data in a post-training stage. We study how the model behaves as it scales. 语言和视觉领域的基石模型通过遵循经验缩放定律不断推进前沿：能力随着数据、参数和计算量的增加而提升。然而，机器人领域却错失了这一机遇。大规模、高质量的数据难以获取，这种稀缺性在很大程度上限制了策略模型的扩展上限。在真正大规模训练下机器人能做到什么程度，这在很大程度上仍是一个未解之谜。我们迈出了回答这一问题的一步。Xiaomi-Robotics-1 将大规模无具身（UMI）预训练与后训练阶段适量的真实机器人数据相结合，并研究了模型在扩展过程中的表现。

Data

数据

Everything Xiaomi-Robotics-1 can do starts from data. For pre-training, we use 100,000 hours of embodiment-free (UMI) trajectories spanning more than 1,700 scenarios (household, commercial premises, industrial sites, and outdoor spaces), covering a diverse range of tasks. We develop a scalable auto-labeling pipeline that first divides trajectories into fixed-length segments and then annotates each segment with language descriptions of scene state transitions. Xiaomi-Robotics-1 的一切能力始于数据。在预训练阶段，我们使用了 10 万小时的无具身（UMI）轨迹，涵盖了超过 1,700 个场景（家庭、商业场所、工业场地和户外空间），涉及多种多样的任务。我们开发了一个可扩展的自动标注流水线，首先将轨迹划分为固定长度的片段，然后用场景状态转换的语言描述对每个片段进行标注。

For post-training, we leverage cross-embodiment datasets containing in-house robot data, filtered open-sourced robot data, and a set of high-quality UMI data. For the in-house data, we collected over 7,200 hours of real-robot data in real homes, covering tasks like tidying a sofa, sorting a shoe cabinet, and putting away kitchenware. The UMI data are manually annotated with temporal segments and instruction prompts, which differ from the auto-labeled state-transition descriptions used in the pre-training data. 在后训练阶段，我们利用了跨具身数据集，其中包含自研机器人数据、经过筛选的开源机器人数据以及一组高质量的 UMI 数据。对于自研数据，我们在真实家庭环境中收集了超过 7,200 小时的真实机器人数据，涵盖了整理沙发、整理鞋柜和收纳厨具等任务。UMI 数据通过人工标注了时间片段和指令提示，这与预训练数据中使用的自动标注状态转换描述有所不同。

Method

方法

Following the training paradigm of LLMs, the training of Xiaomi-Robotics-1 consists of two stages: pre-training and post-training. The first stage learns general representations for action generation from large-scale UMI data, while the post-training stage aligns the model with real robot embodiments and instruction-following capabilities. 遵循大语言模型（LLM）的训练范式，Xiaomi-Robotics-1 的训练分为两个阶段：预训练和后训练。第一阶段从大规模 UMI 数据中学习动作生成的通用表征，而后训练阶段则使模型与真实机器人具身及指令遵循能力对齐。

Pre-training 预训练

Pre-training is about breadth: exposing the model to as much of the real world as possible. We use the embodiment-free UMI data described above, which spans a broad range of environments and tasks. At this scale, manual labeling is infeasible. Thus, we built an automatic annotation pipeline powered by a strong vision-language model. Long videos are split into fixed-length clips, and the VLM describes the state transition of grippers and interacting objects within each clip. The result is a large-scale corpus of real-world manipulation trajectories, each annotated with precise language descriptions. These allow the model to learn action generation that drives the scene toward the state transitions described by the language. 预训练的核心在于广度：让模型尽可能多地接触真实世界。我们使用了上述无具身 UMI 数据，其涵盖了广泛的环境和任务。在如此规模下，人工标注是不可行的。因此，我们构建了一个由强大的视觉语言模型（VLM）驱动的自动标注流水线。长视频被分割成固定长度的片段，VLM 描述每个片段中夹爪和交互物体的状态转换。最终得到的是一个大规模的真实世界操作轨迹语料库，每一条都带有精确的语言描述。这些数据使模型能够学习动作生成，从而驱动场景向语言所描述的状态转换方向发展。

An encouraging finding is that pre-training shows a clean scaling behavior: as data and model size grow, validation action error steadily decreases. 一个令人振奋的发现是，预训练表现出了清晰的缩放行为：随着数据和模型规模的增长，验证动作误差稳步下降。

Post-training 后训练

Post-training aims to align the strong action-generation capabilities acquired from pre-training with real robot embodiments and natural-language instruction following along two axes. Embodiment alignment uses high-quality cross-embodiment real-robot data to map the general action-generation ability onto actual robots. Instruction alignment shifts the model from “generating actions given a description of scene state transitions” to “understanding a natural-language instruction and executing it directly.” 后训练旨在将预训练获得的强大动作生成能力，在两个维度上与真实机器人具身及自然语言指令遵循能力对齐。具身对齐利用高质量的跨具身真实机器人数据，将通用的动作生成能力映射到实际机器人上。指令对齐则将模型从“根据场景状态转换描述生成动作”转变为“理解自然语言指令并直接执行”。

After post-training, Xiaomi-Robotics-1 can be used out-of-the-box to perform a wide range of mobile manipulation tasks in the real world. We evaluate the post-trained model in unseen environments with unseen object instances to understand whether the scaling behaviors from pre-training can transfer to real-robot performance after post-training. 后训练完成后，Xiaomi-Robotics-1 可以开箱即用，在真实世界中执行广泛的移动操作任务。我们在未见过的环境和未见过的物体实例中评估了后训练模型，以了解预训练中的缩放行为是否能在后训练后迁移到真实机器人的性能上。

The answer is yes. As we increase the amount of pre-training data and model size, real-robot success rate rises steadily and predictably. That is, a stronger pre-trained model yields better real-robot performance. The scaling gains show no signs of saturation: the real-robot success rate after post-training keeps improving as the model consumes more data or scales up during pre-training. 答案是肯定的。随着我们增加预训练数据量和模型规模，真实机器人的成功率稳步且可预测地提升。也就是说，更强的预训练模型能带来更好的真实机器人性能。这种缩放带来的收益没有显示出饱和的迹象：随着模型在预训练阶段消耗更多数据或进行扩展，后训练后的真实机器人成功率持续提高。

Applications

应用

After post-training, Xiaomi-Robotics-1 can serve as a strong robot foundation model for downstream applications. We put Xiaomi-Robotics-1 to use in two complementary downstream settings. Efficient adaptation to new tasks specializes the model to brand-new, highly complex real-robot tasks from a few hours of data per task. Simulation benchmarks probe its capabilities in mainstream suites that emphasize generalization. 后训练完成后，Xiaomi-Robotics-1 可以作为下游应用的强大机器人基石模型。我们将 Xiaomi-Robotics-1 应用于两个互补的下游场景。高效的新任务适应能力使模型能够通过每个任务仅几小时的数据，专门处理全新的、高度复杂的真实机器人任务。仿真基准测试则在强调泛化能力的主流套件中探测其能力。

Efficient Adaptation to New Tasks 高效的新任务适应

Xiaomi-Robotics-1 can learn new tasks with high data efficiency. The model picks up tasks like phone packing, printer refilling, laundry loading, and box packing from just a few hours of real-robot demonstrations per task. With an average of under 10 hours of demonstrations per task, it already reaches a 75% overall success rate, nearly doubling the π0.5 baseline (40%) at the same budget; raising the budget to an average of under 40 hours lifts overall success to 85%. Xiaomi-Robotics-1 能够以极高的数据效率学习新任务。模型仅需每个任务几小时的真实机器人演示，即可学会手机装箱、打印机加纸、装载衣物和纸箱打包等任务。在平均每个任务少于 10 小时演示的情况下，它已经达到了 75% 的总体成功率，在相同预算下几乎是 π0.5 基准（40%）的两倍；将预算提高到平均少于 40 小时，总体成功率可提升至 85%。

(Table omitted for brevity) (表格略)

Simulation Benchmarks 仿真基准测试

We evaluate Xiaomi-Robotics-1 on four mainstream simulation benchmarks. It achieves state-of-the-art results on all four benchmarks. The table reports the average success rate and the relative gain over second place. These results show that the generalization and scaling gains of Xiaomi-Robotics-1 carry over to standard simulation evaluation. 我们在四个主流仿真基准测试中评估了 Xiaomi-Robotics-1。它在所有四个基准测试中均取得了最先进（SOTA）的结果。下表报告了平均成功率以及相对于第二名的提升幅度。这些结果表明，Xiaomi-Robotics-1 的泛化能力和缩放收益可以迁移到标准的仿真评估中。

(Table omitted for brevity) (表格略)