DexH13

Source-backed capabilities

• Capability areas: Tactile Dexterity, Robotic grasping, Tactile feedback
• Workflows: End-effector manipulation
• Operating environments: Humanoid robots, manipulation research and tactile-perception experiments
• Target markets: Teams evaluating tactile hand modules and PaXini's wider tactile-perception stack.
• Deployment model: Official PaXini product/store page.
• Tactile interaction sensing

  Research demonstration - 2025-05-22 - PaXini and Shanghai Jiao Tong University researchers collect physical DexH13 tactile signals across four grasp poses and eight objects using 11 arrays and 1,140 three-axis sensing units, then benchmark lossless and lossy compression.

  Source
• Robotic grasping

  Research demonstration - 2025-05-22 - PaXini and Shanghai Jiao Tong University researchers collect physical DexH13 tactile signals across four grasp poses and eight objects using 11 arrays and 1,140 three-axis sensing units, then benchmark lossless and lossy compression.

  Source

• Cross-Hand Latent Representation for Vision-Language-Action Models

  2026-03-10 - Research Publication - XL-VLA evaluates a physical PaXini DexH13 in real-world bimanual xArm experiments spanning ten manipulation tasks. Across ten trials per task, the reported DexH13 mean success rate is 0.78 versus 0.35 for the direct-action baseline; the broader training corpus contains two million state-action pairs across four hand embodiments.

  Source
• TacCompress: A Benchmark for Multi-Point Tactile Data Compression in Dexterous Hand

  2025-05-22 - Research Publication - TacCompress uses a physical PaXini DexH13 supplied at 48 V to collect three-axis tactile stress signals from 11 arrays and 1,140 sensing units across pinch, tripod, cylindrical and spherical grasps on eight objects. The study reports about 200x lossless compression and up to about 1,000x lossy data reduction.

  Source

• PaXini demonstrates tactile robots and dexterous hands at CES

  2026-01-07 - Demonstration - TORA-ONE performed ice-cream preparation, including lever operation, ingredient handling and cup handover; TORA-DOUBLE ONE crossed obstacles and shook hands with visitors. The DexH13 hand mirrored gestures, grasped test tubes and cubes, and turned knobs using 1,140 tactile processing units.

  Source
• PaXiniTech unveils DexH13, the first dual-modal bionic dexterous hand

  2024-12-31 - Product Launch - DexH13 integrates multi-dimensional touch and AI vision to enhance robotic perception. This innovation aims to reshape control boundaries in robotic applications. PaXini is a Shenzhen robotics company focused on tactile sensing and embodied AI.

  Source

Market context

Tactile hand product from PaXini with explicit tactile-unit, tactile-channel and hand-eye camera figures.

• Target environment: Humanoid robots, manipulation research and tactile-perception experiments
• Workflow context: Dexterous grasping, tactile feedback and humanoid hand integration
• Customer context: Teams evaluating tactile hand modules and PaXini's wider tactile-perception stack.
• Deployment model: Official PaXini product/store page.
• Commercial maturity: Commercial product/store page; delivery timing and integration support should be confirmed through PaXini.
• Adoption constraints: Integration depends on tactile-data access, mechanical mounting, control interface and software support.
• Adjacent products: Unitree Dex5-1, Wuji Hand, AIDIN-Hand, RealHand O6

Comparable Systems

• Ability Hand

  PSYONIC - Hand

• DH11

  Pudu Robotics - Dexterous Hand

• PXDex III

  PaXini - Robotic Hand

• ROH-AP002

  OYMotion - Hand

• Torobo Hand

  Tokyo Robotics - Hand

• Allegro Hand V5 Plus

  Wonik Robotics - Dexterous Hand