DEPLOYDatabase

Robot model

RLDX-1

RLDX-1 is a dexterity-first robotics foundation model built for high degree-of-freedom (DoF) robotic hands.

Track RLDX-1Get notified of verified changes
Manufacturer
RLWRLD
Form factor
other
Maturity
pilot
Lifecycle
active

Overview

RLDX-1 is a dexterity-first robotics foundation model built for high degree-of-freedom (DoF) robotic hands. It uses a Multi-Stream Action Transformer (MSAT) architecture with dedicated streams for vision, tactile/torque, and memory, and a robot-specialized VLM fine-tuned on trajectory VQA. RLDX-1 scored 70.6 on RoboCasa Kitchen, outperforming NVIDIA GR00T N1.6 (66.2) using ~20% of the training compute. Deployable across single-arm, dual-arm, and humanoid embodiments.

Verified vs. claimed

Maturity stage
pilot(Small-scale deployment in controlled or trial conditions.)
Verified deployments
None on file
Sources on file
2 sources, view all

No verified deployments on file. DEPLOY tracks independently-sourced deployment evidence separately from manufacturer capability claims.

Data & sources

News coverage

1

Web sources

1

2 sources backing this record.View all →

Pricing

No verified price is on record for RLDX-1. Physical-AI systems are often sold through enterprise contracts or operated as a service rather than at a public list price.

How DEPLOY tracks pricing

RLDX-1 on the deployment map

Where RLDX-1 is verified operating. Explore the deployment map by place and type.

Safety record

No incidents on record for RLDX-1.

Only active incidents are counted. Retracted incidents are excluded from this summary but remain reachable at their canonical URLs.

Sources (2)

  1. RLWRLD releases RLDX-1, a dexterity-first foundation model for robot hands · https://www.therobotreport.com/rlwrld-releases-rldx-1-a-dexterity-first-foundation-model-for-robot-hands/
  2. RLWRLD Debuts Robotics Foundation Model RLDX-1 at NVIDIA GTC Taipei 2026 · https://theaiinsider.tech/2026/06/02/rlwrld-debuts-robotics-foundation-model-rldx-1-at-nvidia-gtc-taipei-2026/

Common questions

What is RLDX-1?
RLDX-1 is a dexterity-first robotics foundation model built for high degree-of-freedom (DoF) robotic hands. It uses a Multi-Stream Action Transformer (MSAT) architecture with dedicated streams for vision, tactile/torque, and memory, and a robot-specialized VLM fine-tuned on trajectory VQA. RLDX-1 scored 70.6 on RoboCasa Kitchen, outperforming NVIDIA GR00T N1.6 (66.2) using ~20% of the training compute. Deployable across single-arm, dual-arm, and humanoid embodiments.
How much does RLDX-1 cost?
RLDX-1's price is not publicly disclosed. DEPLOY has no verified price on record for RLDX-1 from RLWRLD. Physical-AI systems like this are often sold through enterprise contracts or operated as a service rather than at a public list price; check the manufacturer for the latest.
Is RLDX-1 actually deployed in the real world?
RLDX-1 is at the pilot stage: a trial is reported, but no deployment is yet verified at a named site on the DEPLOY registry. Manufacturer capability claims should be read as claims pending independent verification.
Who makes RLDX-1?
RLDX-1 is made by RLWRLD, based in Seoul, South Korea, founded in 2024.
Can you buy RLDX-1?
RLDX-1 is in pilot deployments with named customers and is not yet broadly for sale.
Methodology: Unreviewed · 2 sources (no primary) · last reviewed 2026-07-16

Verification posture

Unreviewed

Low confidence

Review state

Stable

Last reviewed 2026-07-16

Maturity + lifecycle

Maturity stage: pilot

Lifecycle: active

Architectural position

Cohort: other

Sources by quality tier

1
secondary-trade-publication
Trade publication
1
unclassified
Unclassified source

The framework is documented at /methodology. Corrections at /corrections. Reviewer: DEPLOY editorial team.

Methodology surface for RLDX-1.