Zhicheng AI · Towards AGI
中 / EN Contact sales
EMBODIED AI · PHYSICAL INTELLIGENCE

Robots that understand the physical world

Zhicheng AI builds the Chengling World Model, an end-to-end neural network that lets robots predict outcomes before they act — moving from seeing the world to understanding, predicting and changing it.

Explore products Book a demo
TR-series humanoid robot seated in a thinking pose, its visor glowing blue
2024.03
Founded in Hangzhou
1 × 2
One brain, two robot forms
53 DoF
TR6 humanoid degrees of freedom
50 kg
XR1 dual-arm payload
ABOUT ZHICHENG AI

The endgame of general AI isn’t a smarter chat window. It’s a robot that works and lives alongside us.

Z Robotics Inc (“Zhicheng AI”) was founded in March 2024 to focus on embodied intelligence. We design and train an AI-native physical-world model end to end, giving robots stronger generalization in real environments.

VISIONBring general AI into the physical world, serving every person and organization
MISSIONLead the next generation of general AI with platforms and robots that understand the physical world
PRODUCTS

One brain, two forms

TR and XR share our in-house Chengling physical-intelligence brain. Form follows the task while intelligence transfers: XR serves industry today, TR is built for human environments tomorrow.

TR SERIES · TRANSFORM ROBOT
New for 2026
TR6 bipedal humanoid robot, standing front view

TR6 General-Purpose Humanoid

Details →

A bionic bipedal design with hybrid series-parallel actuation, built for spaces made for people: stairs, thresholds, narrow corridors.

176 cmHeight · 62 kg
53 DoFDegrees of freedom
8 hRuntime · 2 m/s walk
XR SERIES · EXTENDED ROBOTICS
Available now
XR1 wheeled robot with two arms, lifting torso and omnidirectional base

XR1 Wheeled Dual-Arm Robot

Details →

Dual-arm manipulation, omnidirectional mobility and a lifting torso. Modular and ready out of the box for assembly, inspection and handling.

50 kgDual-arm payload
25 DoFDegrees of freedom
24/7Autonomous battery swap
CHENGLING WORLD MODEL

Rehearse the outcome
before making a move

The hard part isn’t seeing — it’s predicting. Chengling models state, causality and the consequences of actions together, cutting costly trial and error in the real world.

01 · INPUT
Perceive
Language: tasks, constraints
Vision: RGB, depth
Touch: tactile, force
Body: joints, trajectories
02 · WORLD STATE
Understand
Space and objects
Relations between objects
Dynamics
Task intent
03 · PREDICT
Predict
What happens next
What if I do A
What can be manipulated
Where uncertainty remains
04 · ACT
Act
Position and direction
Frequency and speed
Force control, feedback
Millisecond closed loop
Feedback · continuous correction and replanning
REACTIVEReactive policy (typical VLA approach)
Decisions based mainly on the current observation
Errors compound over long tasks
Real-world trial and error is costly
PREDICTIVEPredictive closed loop (Chengling)
Explicitly models the 3D world state
Predicts the consequences of an action before acting
Corrects from feedback — suited to long, contact-rich tasks
IN THE REAL WORLD

In the lab: capability.
In the real world:
mastery.

Light changes, objects get occluded, tasks change mid-stream. XR1 grasps, hands over and collaborates in the unstructured settings of trade shows and production lines.

XR1 handing an item to a visitor at a trade show
APPLICATIONS

Start where value is proven

Robots aren’t here to replace people, but to free them from dangerous, repetitive and draining work.

Smart manufacturing
Flexible assembly and QC for high-mix, low-volume lines
Biochemical testing
Lab-bench work with strict precision and safety demands
Logistics
High-throughput sorting, handling and transfer
Research & education
An embodied-AI research platform with ROS support
Security patrol
Autonomous inspection and anomaly detection across sites
Reception & tours
Greeting, wayfinding and interactive exhibit tours
HOW WE WORK Platform sales Enterprise solutions Co-development Discuss your use case →
RESEARCH

Open research,
verifiable progress

From spatial visual perception to cross-embodiment manipulation, we publish our core methods.

All papers →
  1. 2025.10EBT-Policy: Energy Unlocks Emergent Physical Reasoning CapabilitiesarXiv:2510.27545 ↗
  2. 2025.09Tenma: Robust Cross-Embodiment Robot Manipulation with Diffusion TransformerarXiv:2509.11865 ↗
  3. 2025.05Spatial RoboGrasp: Generalized Robotic Grasping Control PolicyarXiv:2505.20814 ↗
  4. 2024.11Spatially Visual Perception for End-to-End Robotic LearningarXiv:2411.17458 ↗
  5. 2024.09Generalized Robot Learning FrameworkarXiv:2409.12061 ↗
NEWS

Latest

All news →
TR-series humanoid robot at the Zhicheng AI booth 2026.09.23 · EXHIBITION XR1 and TR6 at the 5th Global Digital Trade Expo, booth 4-T088
CCTV news footage of a Zhicheng AI robot arm at work 2025.05.20 · PRESS Zhicheng AI robots featured on CCTV News Coverage:Xinhua ↗People’s Daily ↗gov.cn ↗
Award ceremony of the 2025 Global Digital Trade Innovation Competition robotics track 2025.09.25 · COMPETITION Zhicheng AI wins silver in the robotics track of the 2025 Global Digital Trade Innovation Competition Coverage:Guangming Online ↗
HONORS 2025.02 · Yuhang District Education–Tech–Talent Integration Award 2025.11 · TR5 featured in “2025 AI+ Industry Ecosystem Paradigms”
CUSTOMERS & PARTNERS

Working with industry and research partners

Industry
Huawei logo
JD.com logo
State Grid logo
FAW-Volkswagen logo
NIO logo
Oulin logo
Manycore Tech logo
Cainiao logo
WuBa Intelligent Tech logo
Amazon logo
Research
Peking University logo
Stanford University logo
Tsinghua University logo
Harvard University logo
Zhejiang University logo

Bring us your hardest
real-world problem

ADDRESSRoom 908, Building A2, Cainiao Smart Valley Industrial Park Phase I, Yuhang District, Hangzhou
Zhicheng AI WeChat official account QR code Follow Zhicheng AI on WeChat
Zhicheng AI · Towards AGI
中文
EMBODIED AI · PHYSICAL INTELLIGENCE

Robots that understand the physical world

Our Chengling World Model lets robots predict outcomes before they act — from seeing the world to understanding, predicting and changing it.

Products Book a demo
TR-series humanoid robot seated in a thinking pose
2024.03Founded in Hangzhou
1 × 2One brain, two forms
53 DoFTR6 degrees of freedom
50 kgXR1 dual-arm payload
ABOUT ZHICHENG AI

The endgame of general AI isn’t a smarter chat window. It’s a robot that works and lives alongside us.

Founded in Hangzhou in March 2024, Zhicheng AI designs and trains an AI-native physical-world model end to end, giving robots stronger generalization in real environments.

PRODUCTS

One brain, two forms

TR SERIESTR6 bipedal humanoid robot
TR6 General-Purpose Humanoid176 cm · 62 kg · 53 DoF · 8 h runtimeDetails →
XR SERIESXR1 wheeled dual-arm robot
XR1 Wheeled Dual-Arm Robot90–190 cm · 160 kg · 25 DoF · 50 kg dual-armDetails →
CHENGLING WORLD MODEL

Rehearse the outcome before making a move

Chengling models state, causality and the consequences of actions together, cutting costly trial and error in the real world.

  1. 01
    PerceiveLanguage · RGB-D vision · touch and force · joint trajectories
  2. 02
    UnderstandSpace · object relations · dynamics · task intent
  3. 03
    PredictFuture states · consequences · what can be manipulated · uncertainty
  4. 04
    ActPosition · direction · frequency · speed · force feedback
↺ Feedback · continuous correction and replanning
XR1 interacting with a visitor at a trade show

In the lab: capability.
In the real world: mastery.

APPLICATIONS

Start where value is proven

ManufacturingFlexible assembly and QC
Biochemical testingPrecise lab-bench work
LogisticsSorting, handling, transfer
Research & educationROS development platform
Security patrolAutonomous site inspection
Reception & toursGreeting and guided tours
NEWS
Humanoid robot at the Zhicheng AI booth 2026.09.23 · EXHIBITION XR1 and TR6 at the 5th Global Digital Trade Expo, booth 4-T088
2025.05.20 · Zhicheng AI robots featured on CCTV NewsCoverage:Xinhua ↗People’s Daily ↗gov.cn ↗
2025.09.25 · Silver award, robotics track, 2025 Global Digital Trade Innovation CompetitionCoverage:Guangming Online ↗
Paper: Tenma — cross-embodiment robot manipulation with a Diffusion Transformer
CUSTOMERS & PARTNERS

Working with industry and research partners

Industry
Huawei logo
JD.com logo
State Grid logo
FAW-Volkswagen logo
NIO logo
Oulin logo
Manycore Tech logo
Cainiao logo
WuBa Intelligent Tech logo
Amazon logo
Research
Peking University logo
Stanford University logo
Tsinghua University logo
Harvard University logo
Zhejiang University logo

Bring us your hardest real-world problem

customer@zhicheng-ai.com
+86 186 5719 6422
Room 908, Building A2, Cainiao Smart Valley, Yuhang District, Hangzhou
Zhicheng AI WeChat official account QR code Long-press to scan and follow Zhicheng AI on WeChat
Book a demo