XiaomiChina2026
Xiaomi-Robotics-1
An AI model: a vision-language-action model (VLA) takes camera images and an instruction and turns them into movements for a robot. This page describes the software, not a robot.
Robotsnot stated
Key figures as stated by the developer
5 of 6 stated · 1 gap-
Model type C
vision-language-action -
Controls A
arms and hands -
Inputs A
cameralanguagerobot state - Access B open download
- Parameters not stated
- Base model A Qwen3-VL-4B-Instruct
How to read the figures
- 14kgA30.9lbA
- The letter in the grey square after a figure points to the source the figure came from. Robots with several sources carry several letters.
- 74.5lb
- Imperial figures are our own conversion unless the manufacturer states an imperial figure itself. The source letter always belongs to the metric figure — a conversion has no source of its own.
Versions and notes in the key figures (6)
Model type
Developer wording: "We present Xiaomi-Robotics-1, a foundational vision-language-action (VLA) model".
Controls
"XR-1 is a Vision-Language-Action (VLA) foundation model for mobile manipulation".
Inputs
"The runtime client accepts three camera images, a natural-language instruction, and the current robot state".
Access
The repository says the model has been "open-sourced on Hugging Face".
Parameters
Note The released checkpoint is named "Xiaomi-Robotics-1-5B".
Base model
"The Qwen3-VL backbone is loaded from [Qwen/Qwen3-VL-4B-Instruct]".
Full specifications
7 of 12 fields stated| Field | Value | Source and caveats |
|---|---|---|
| Input and output | ||
| Model type | vision-language-action | C Developer wording: "We present Xiaomi-Robotics-1, a foundational vision-language-action (VLA) model". |
| Inputs | camera language robot state | A "The runtime client accepts three camera images, a natural-language instruction, and the current robot state". |
| Controls | arms and hands | A "XR-1 is a Vision-Language-Action (VLA) foundation model for mobile manipulation". |
| Degrees of freedom controlled | not stated | |
| Size and speed | ||
| Parameters | not stated | Note The released checkpoint is named "Xiaomi-Robotics-1-5B". A |
| Control rate | not stated | |
| Where it runs and what it is built on | ||
| Runs where | not stated | Note The repository says the model is "optimized for consumer GPUs". B |
| Base model | Qwen3-VL-4B-Instruct | A "The Qwen3-VL backbone is loaded from [Qwen/Qwen3-VL-4B-Instruct]". |
| Openness | ||
| Access | open download | B The repository says the model has been "open-sourced on Hugging Face". |
| License | Apache 2.0 | A The Hugging Face model card says "license: apache-2.0"; the repository names the "Apache License 2.0". |
| The developer’s own claims | ||
| Learns new tasks | needs retraining | D Developer claim: "Xiaomi-Robotics-1 can learn new tasks with high data efficiency." New tasks are learned "from just a few hours of real-robot demonstrations per task"; the paper describes "efficiently adapting to novel downstream tasks with minimal fine-tuning data". |
| Works across robots | not stated | Note The model is "post-trained on over 10K hours of cross-embodiment data". A Yes only if the developer itself says the model runs on different robot models, for example robots from two manufacturers or two of one manufacturer’s models. Several units of the same robot do not count. |
Notes
- Xiaomi calls Xiaomi-Robotics-1 a vision-language-action model; the project page is dated 16 July 2026.
- Xiaomi describes it as a ready-to-use robot foundation model trained on over 100K hours of real-world manipulation trajectories.
Sources
-
A
Manufacturer product page. huggingface.
co/ · RetrievedXiaomiRobotics/ Xiaomi-Robotics-1-5B -
B
github.
com/ · RetrievedXiaomiRobotics/ Xiaomi-Robotics-1 -
C
arxiv.
org/ · Retrievedabs/ 2607. 15330 -
D
robotics.
xiaomi. · Retrievedcom/ xiaomi-robotics-1. html
One line per source, not per figure: a source has one retrieval date. The letter beside a figure points to the line it came from.