1X TechnologiesUnited States2025

Redwood

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.

RobotsNEOby 1X Technologies AEVEby 1X Technologies A

Key figures as stated by the developer

4 of 6 stated · 2 gaps

Model type ⁠A vision-language-action
Controls ⁠A arms and hands whole body walking
Inputs ⁠A camera language robot state
Access not stated
Parameters ⁠A 160,000,000
Base model not stated

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.

The four states

Versions and notes in the key figures (6)

Model type

Developer wording: "Redwood is among the first VLAs to control locomotion jointly with manipulation".

Controls

Announcement: "Redwood predicts not only the arm and hand commands, but also walking, manipulation, and pelvis pose commands simultaneously."

Inputs

Announcement: "Redwood fuses pre-trained language embeddings, vision tokens from a pre-trained vision transformer, and proprioception embeddings from a sequence of joint positions and joint applied forces."

Access

Note 1X Technologies says "Redwood, 1X’s breakthrough AI model that we will be deploying to homes".

Parameters

Announcement: "despite its small size (160M parameters)".

Base model

Note Pretrained parts: "Redwood fuses pre-trained language embeddings, vision tokens from a pre-trained vision transformer, and proprioception embeddings from a sequence of joint positions and joint applied forces."

Full specifications

7 of 12 fields stated

The letter points to the source at the foot of the page. Fields without a figure are dashed: the developer does not state them — that is not the same as zero, and not the same as no.
Field	Value	Source and caveats
Input and output
Model type	`vision-language-action`	A Developer wording: "Redwood is among the first VLAs to control locomotion jointly with manipulation".
Inputs	`camera` `language` `robot state`	A Announcement: "Redwood fuses pre-trained language embeddings, vision tokens from a pre-trained vision transformer, and proprioception embeddings from a sequence of joint positions and joint applied forces."
Controls	`arms and hands` `whole body` `walking`	A Announcement: "Redwood predicts not only the arm and hand commands, but also walking, manipulation, and pelvis pose commands simultaneously."
Degrees of freedom controlled	not stated
Size and speed
Parameters	160,000,000	A Announcement: "despite its small size (160M parameters)".
Control rate	approximately 5Hz	A Rate of the model on the robot's GPU, not of the motor commands: "Redwood is a 160M parameter transformer model that runs on NEO’s onboard GPU at around 5hz."
Where it runs and what it is built on
Runs where	on the robot	A "Redwood is compute-efficient and runs fully on NEO’s onboard embedded GPU."
Base model	not stated	Note Pretrained parts: "Redwood fuses pre-trained language embeddings, vision tokens from a pre-trained vision transformer, and proprioception embeddings from a sequence of joint positions and joint applied forces." ⁠A
Openness
Access	not stated	Note 1X Technologies says "Redwood, 1X’s breakthrough AI model that we will be deploying to homes". ⁠A
License	not stated
The developer’s own claims
Learns new tasks	not stated	Note On generalization the announcement says: "picking up never-before-seen objects in unfamiliar locations". ⁠A
Works across robots	yes	A Developer claim, for two different 1X robots (EVE and NEO): "In order to power autonomy on both EVE and NEO platforms". "We decode this representation into EVE or NEO actions using a diffusion policy" 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

1X Technologies describes Redwood as a vision-language transformer for the humanoid form factor, which it will deploy to homes, and which does mobile manipulation end to end.
1X Technologies published the Redwood announcement on 10 June 2025.

Sources

A Manufacturer product page. www.1x.tech/discover/redwood-ai · Retrieved 2026-10-14

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.