Vega 1 Pro
About
About Vega 1 Pro
Vega 1 Pro is Dexmate's wheeled dual-arm robot for physical-AI research, data collection and application development. Two seven-axis arms and an adjustable torso sit on an omnidirectional base, allowing the robot to reach across benches, work near the floor and move between task locations. End effectors are ordered separately; teams can choose Dexmate hands or grippers, or fit their own tools using the published interfaces.
The platform includes Python, C++ and Rust APIs, simulation models and tools for recording demonstrations through VR or exoskeleton teleoperation. It ships with position control, sensors and safety checks, while users build or train the task behaviour. The current Vega 1 Pro is in production and offered directly to developers and researchers, with optional force/torque sensing for compliant control.
Capabilities
Source-backed capabilities
- Operating environments: Research and application-development spaces
- Control modes: User Implemented, Teleoperated
- Whole-body loco-manipulation
Claimed - Prototype pick, carry and place workflows using user-developed skills on Vega 1 Pro. - environment: Research and application-development spaces
Source - Remote and shared control
Claimed - Control Vega 1 Pro through VR or an exoskeleton to record demonstration trajectories. - environment: Research and application-development spaces
Source
Observed use
Exact-product configurations are shown without inferring supply, purchase, customer status, standard fit, deployment or partnership.
- Sharpa Wave
Research Experiment - Mounted - Physical - FTP-1 evaluates Dexmate Vega with two mounted Sharpa Wave tactile hands in a contact-rich manipulation research configuration.
Source - OYMotion six-DoF dexterous hands
Research Experiment - Mounted - Physical - ResFiT evaluates Dexmate Vega with two mounted six-DoF OYMotion hands for ball picking and package handover.
Source - ZED stereo camera
Research Experiment - Mounted - Physical - The ResFiT Vega research configuration uses a head-mounted ZED camera providing two image streams. The exact camera model is not named.
Source
Supply
Supply / Components
- Dexmate Vega
Jetson Compute Module - Compute & Control - NVIDIA - Dexmate Vega integrates NVIDIA AGX Thor compute - source: dexmate.ai
Source
Research
- FACT: A Multimodal Force-Action-Track Foundation Model for Human-to-Robot Dexterity
2026-10-13 - Research Publication - FACT combines visual planning with force-aware action refinement for dexterous manipulation. Its ten-task physical evaluation uses YAM Ultra arms and Sharpa Wave hands; robot pretraining also reuses the T-Rex collection from a Dexmate Vega-1.
Source - What Should Embodied Agents Remember? Benchmarking and Evaluating Memory Modality for Embodied Agents
2026-10-08 - Research Publication - MeMoREval compares memory modalities for embodied agents across 53 tasks and four simulated indoor scenes, measuring task success and memory inference costs. Two physical Dexmate Vega demonstrations test how stored experiences support later requests.
Source - InterMimicGen: Scaling Humanoid Loco-Manipulation through Self-Evolving Motion Imitation
2026-10-05 - Research Publication - InterMimicGen couples humanoid motion references with a learned tracker in an iterative data-generation loop. It retargets human-object interactions to several embodiments and demonstrates selected motions on physical G1, Booster K1 and Dexmate Vega-1 Pro robots.
Source - DexTacWAM: A Visuo-Tactile World-Action Model for Dexterous Manipulation
2026-09-21 - Research Publication - DexTacWAM extends a video world-action model with per-fingertip tactile prediction and a hand-aware tactile compressor. Across six contact-rich tasks on a bimanual Dexmate and Sharpa setup, it averages 70.6 against 38.0 for the strongest baseline.
Source - ReVeal: A Reconstruction-Aware Real-to-Sim Framework for VLA Policy Evaluation
2026-09-20 - Research Publication - ReVeal assesses reconstructed environments for vision-language-action policy evaluation using visual and planar geometry metrics. Eight matched physical and simulated Dexmate Vega tasks test how reconstruction fidelity affects policy success.
Source - Learning Dexterous Manipulation Using Contact Wrench Guidance From Human Demonstration
2026-06-22 - Research Publication - CHORD learns contact-rich dexterous manipulation from human demonstrations by matching object-centric contact wrenches. It evaluates a large simulated task library and transfers selected policies to a fixed-base Dexmate robot with two Sharpa hands.
Source - MagicSim: A Unified Infrastructure for Executable Embodied Interaction
2026-06-16 - Research Publication - MagicSim proposes a shared runtime for simulated robot control, skills and planning, combining executable robot embodiments with multiple drivers. Its registry includes Unitree G1 variants and a Vega-1P mobile manipulator with Sharpa Wave hands.
Source - T-Rex: Tactile-Reactive Dexterous Manipulation
2026-06-15 - Research Publication - T-Rex evaluates two physical 22-DoF Sharpa Wave hands with per-fingertip tactile deformation and six-axis wrench sensing. The work introduces 100 hours of tactile-rich robot data and reports more than 30% higher average success than the strongest baseline on 12 force-sensitive and deformable-object tasks.
Source - Contrastive Action-Image Pre-training for Visuomotor Control
2026-06-15 - Research Publication - CAIP learns action-centred visual representations from 32,041 hours of egocentric human video and 88 hours of robot manipulation data. Physical evaluation uses a Dexmate Vega with two 22-DoF Sharpa Wave hands and reports more than 30% gains on folding, pouring and fine-grained manipulation tasks.
Source - FTP-1: A Generalist Foundation Tactile Policy Across Tactile Sensors for Contact-Rich Manipulation
2026-06-11 - Research Publication - FTP-1 trains a generalist tactile policy across heterogeneous sensors and embodiments using about 3,000 hours from 26 data sources. Physical evaluations include Sharpa North and a Sharpa-equipped Dexmate platform, alongside a separate simulated benchmark.
Source - PGDG: Physically Grounded Data Generation for Robust Bimanual Policy Learning from a Single Demonstration
2026-05-20 - Research Publication - PGDG expands one bimanual demonstration into physically grounded recovery data, combining rollout sampling, diversity curation and corrective action relabeling. Four tasks test simulated learning and zero-shot transfer to a Dexmate Vega.
Source - Residual Off-Policy RL for Finetuning Behavior Cloning Policies
2025-09-23 - Research Publication - ResFiT improves a frozen behavior-cloning policy with lightweight residual corrections learned by off-policy reinforcement learning. Its physical experiments use a Dexmate Vega with OyMotion hands for ball picking and package handover, alongside separate simulated manipulation studies.
Source - Dexmate Vega robot-description and simulation models
Research Publication - Dexmate's model collection provides Vega robot descriptions, collision models and named hand/gripper configurations for simulation and planning.
Source
Resources
Curated durable documents and official technical references.
- Vega 1 Pro full specification sheet
PDF · Vega 1 Pro · Dexmate
Source
Context
Market context
Vega is a bimanual mobile manipulator combining dual arms with an omnidirectional base.
- Target environment: Factories, warehouses, labs, service sites and controlled indoor environments where wheeled manipulation can operate safely.
- Workflow context: Autonomous navigation, object handling, dual-arm or arm-based manipulation, teleoperation and facility task execution.
- Customer context: Industrial operators, logistics teams, research labs and service organizations evaluating mobile manipulation.
- Deployment model: Commercial or available product positioned for customer use, integration or developer adoption.
- Commercial maturity: Publicly presented as commercial or available, with maturity depending on deployment scale and support coverage.
- Adoption constraints: Adoption depends on site fit, manipulation reliability, safety, navigation, integration effort and task economics.
- Market position: Vega sits in the mobile-manipulation lane, using mobility plus arms to pursue practical physical tasks.
- Adjacent products: Kinisi 01, DexForce W1, Stretch RE2, Quanta X2