Skip to content
Trending eventWatching

JD Logistics × Zhiyuan Robotics: Embodied Robot G2 Max Applied in Zhi Wang Warehouse

7 reports2 reporting sourcesUpdated 3 days ago

Overview

Report summary

Zhiyuan Robotics has deployed the G2 Max, an industrial-grade embodied robot, in JD Logistics' Zhi Wang warehouse for inbound handling tasks. The robot can carry up to 50kg with high precision, achieving over 99% operational success rate. This marks a real-world application of embodied intelligence, with JD Logistics emphasizing the importance of industrial scenarios for testing and refining such technologies.

From Chinese robotics — Deployment discovery

Event evidence and corrections

0 attributed source owners. Ownership does not establish independent confirmation. Quantities are reported separately and are never added together.

Reported quantity · robot_trajectories: 780000 other · trajectory count · Differing source assertions
Supporting report

“英伟达去年3月公布过一组数据:用它的工具,11个小时就能生成78万条机器人合成动作轨迹,相当于人类示范6500个小时的数据量。”

Exact source · revision 2

Source owner not reported

Supporting report

“英伟达去年3月公布过一组数据:用它的工具,11个小时就能生成78万条机器人合成动作轨迹,相当于人类示范6500个小时的数据量。”

Exact source · revision 2

Source owner not reported

Reported quantity · robot_trajectories: 6500 hours · Basis not reported · Differing source assertions
Supporting report

“英伟达去年3月公布过一组数据:用它的工具,11个小时就能生成78万条机器人合成动作轨迹,相当于人类示范6500个小时的数据量。”

Exact source · revision 2

Source owner not reported

Supporting report

“英伟达去年3月公布过一组数据:用它的工具,11个小时就能生成78万条机器人合成动作轨迹,相当于人类示范6500个小时的数据量。”

Exact source · revision 2

Source owner not reported

Reported quantity · Reported success rate: 100 percent · Basis not reported · Differing source assertions
Supporting report

“宇树科技创始人王兴兴8月在世界机器人大会上就讲过:在固定场景里,经过充分的数据采集和专项训练,机器人完成任务的成功率几乎能到100%;可只要换一件物品,或者把环境稍微挪一挪,成功率就会大幅下滑。”

Exact source · revision 2

Source owner not reported

Supporting report

“宇树科技创始人王兴兴8月在世界机器人大会上就讲过:在固定场景里,经过充分的数据采集和专项训练,机器人完成任务的成功率几乎能到100%;可只要换一件物品,或者把环境稍微挪一挪,成功率就会大幅下滑。”

Exact source · revision 2

Source owner not reported

Reported quantity · robot_cost: 10000 other · Basis not reported · Differing source assertions
Supporting report

“早在2024年6月的特斯拉股东大会上,他就算过一笔大账:全世界未来每年可能要造10亿台人形机器人,特斯拉的份额可能远超10%;规模足够大时,每台成本可降到1万美元左右,卖2万美元。”

Exact source · revision 2

Source owner not reported

Supporting report

“早在2024年6月的特斯拉股东大会上,他就算过一笔大账:全世界未来每年可能要造10亿台人形机器人,特斯拉的份额可能远超10%;规模足够大时,每台成本可降到1万美元左右,卖2万美元。”

Exact source · revision 2

Source owner not reported

Reported quantity · robot_cost: 20000 other · Basis not reported · Differing source assertions
Supporting report

“早在2024年6月的特斯拉股东大会上,他就算过一笔大账:全世界未来每年可能要造10亿台人形机器人,特斯拉的份额可能远超10%;规模足够大时,每台成本可降到1万美元左右,卖2万美元。”

Exact source · revision 2

Source owner not reported

Supporting report

“早在2024年6月的特斯拉股东大会上,他就算过一笔大账:全世界未来每年可能要造10亿台人形机器人,特斯拉的份额可能远超10%;规模足够大时,每台成本可降到1万美元左右,卖2万美元。”

Exact source · revision 2

Source owner not reported

Reported quantity · robot_production: 1000000000 robots · Basis not reported
Supporting report

“早在2024年6月的特斯拉股东大会上,他就算过一笔大账:全世界未来每年可能要造10亿台人形机器人,特斯拉的份额可能远超10%;规模足够大时,每台成本可降到1万美元左右,卖2万美元。”

Exact source · revision 2

Source owner not reported

Supporting report

“早在2024年6月的特斯拉股东大会上,他就算过一笔大账:全世界未来每年可能要造10亿台人形机器人,特斯拉的份额可能远超10%;规模足够大时,每台成本可降到1万美元左右,卖2万美元。”

Exact source · revision 2

Source owner not reported

Reported quantity · load capacity: 20 other · Basis not reported · Differing source assertions
Supporting report

“作为行业率先实现长程、多任务作业的重载具身机器人,其单臂负载20kg、双臂峰值负载50kg,手臂重复定位精度≤±0.5mm,既“搬得动”也“放得稳”,载重与精细作业兼得;依托高精度腰部俯仰关节,G2 Max可主动调整机身姿态,适配原有工业产线,背后无需预留1.5米绕行空间,可持续、精准地料箱码放作业,每小时可实现拆垛80箱、码垛75箱,整体作业成功率超99%。”

Exact source · revision 2

Source owner not reported

Reported quantity · load capacity: 50 other · Basis not reported · Differing source assertions
Supporting report

“作为行业率先实现长程、多任务作业的重载具身机器人,其单臂负载20kg、双臂峰值负载50kg,手臂重复定位精度≤±0.5mm,既“搬得动”也“放得稳”,载重与精细作业兼得;依托高精度腰部俯仰关节,G2 Max可主动调整机身姿态,适配原有工业产线,背后无需预留1.5米绕行空间,可持续、精准地料箱码放作业,每小时可实现拆垛80箱、码垛75箱,整体作业成功率超99%。”

Exact source · revision 2

Source owner not reported

Reported quantity · repeatability accuracy: 0.5 other · Basis not reported
Supporting report

“作为行业率先实现长程、多任务作业的重载具身机器人,其单臂负载20kg、双臂峰值负载50kg,手臂重复定位精度≤±0.5mm,既“搬得动”也“放得稳”,载重与精细作业兼得;依托高精度腰部俯仰关节,G2 Max可主动调整机身姿态,适配原有工业产线,背后无需预留1.5米绕行空间,可持续、精准地料箱码放作业,每小时可实现拆垛80箱、码垛75箱,整体作业成功率超99%。”

Exact source · revision 2

Source owner not reported

Reported quantity · unloading rate: 80 other · Basis not reported
Supporting report

“作为行业率先实现长程、多任务作业的重载具身机器人,其单臂负载20kg、双臂峰值负载50kg,手臂重复定位精度≤±0.5mm,既“搬得动”也“放得稳”,载重与精细作业兼得;依托高精度腰部俯仰关节,G2 Max可主动调整机身姿态,适配原有工业产线,背后无需预留1.5米绕行空间,可持续、精准地料箱码放作业,每小时可实现拆垛80箱、码垛75箱,整体作业成功率超99%。”

Exact source · revision 2

Source owner not reported

Reported quantity · loading rate: 75 other · Basis not reported
Supporting report

“作为行业率先实现长程、多任务作业的重载具身机器人,其单臂负载20kg、双臂峰值负载50kg,手臂重复定位精度≤±0.5mm,既“搬得动”也“放得稳”,载重与精细作业兼得;依托高精度腰部俯仰关节,G2 Max可主动调整机身姿态,适配原有工业产线,背后无需预留1.5米绕行空间,可持续、精准地料箱码放作业,每小时可实现拆垛80箱、码垛75箱,整体作业成功率超99%。”

Exact source · revision 2

Source owner not reported

Reported quantity · Reported success rate: 99 percent · Basis not reported · Differing source assertions
Supporting report

“作为行业率先实现长程、多任务作业的重载具身机器人,其单臂负载20kg、双臂峰值负载50kg,手臂重复定位精度≤±0.5mm,既“搬得动”也“放得稳”,载重与精细作业兼得;依托高精度腰部俯仰关节,G2 Max可主动调整机身姿态,适配原有工业产线,背后无需预留1.5米绕行空间,可持续、精准地料箱码放作业,每小时可实现拆垛80箱、码垛75箱,整体作业成功率超99%。”

Exact source · revision 2

Source owner not reported

Reported quantity · storage efficiency: 6 other · Basis not reported
Supporting report

“其最大化利用12米的空间,存储效率较传统仓提升6倍,数百台机器人同时工作,出库拣选效率提升3倍。”

Exact source · revision 2

Source owner not reported

Reported quantity · picking efficiency: 3 other · Basis not reported
Supporting report

“其最大化利用12米的空间,存储效率较传统仓提升6倍,数百台机器人同时工作,出库拣选效率提升3倍。”

Exact source · revision 2

Source owner not reported

Reported quantity · assembly_time_per_unit: 2.5 other · Basis not reported · Differing source assertions
Supporting report

“智元机器人2.5分钟造一台“小伙伴”,还学会“无师自通”了”

Exact source · revision 2

Source owner not reported

Reported quantity · assembly_time_per_unit: 2.5 other · Basis not reported · Differing source assertions
Supporting report

“智元发布“造机器人”的机器人,节奏稳不瞌睡2.5分钟组装一台“同伴” 作为关节组装线的“精密搬运工”,G2 Max负责将加工完成的肩关节壳体、髋关节总成等四肢关节配件,从堆料区精准送达总装工位。”

Exact source · revision 2

Source owner not reported

Reported quantity · load_capacity_per_arm: 20 other · Basis not reported · Differing source assertions
Supporting report

“精灵G2 Max单臂负载20公斤,双臂峰值负载可达50公斤,手臂重复定位精度控制在0.5毫米以内,是行业内率先实现长程、多任务作业的重载具身机器人。”

Exact source · revision 2

Source owner not reported

Reported quantity · load_capacity_per_arm: 50 other · Basis not reported · Differing source assertions
Supporting report

“精灵G2 Max单臂负载20公斤,双臂峰值负载可达50公斤,手臂重复定位精度控制在0.5毫米以内,是行业内率先实现长程、多任务作业的重载具身机器人。”

Exact source · revision 2

Source owner not reported

Reported quantity · repeatability_accuracy: 0.5 other · Basis not reported · Differing source assertions
Supporting report

“精灵G2 Max单臂负载20公斤,双臂峰值负载可达50公斤,手臂重复定位精度控制在0.5毫米以内,是行业内率先实现长程、多任务作业的重载具身机器人。”

Exact source · revision 2

Source owner not reported

Supporting report

“精灵G2 Max单臂负载20公斤,双臂峰值负载可达50公斤,手臂重复定位精度控制在0.5毫米以内,是行业内率先实现长程、多任务作业的重载具身机器人。”

Exact source · revision 2

Source owner not reported

Reported quantity · disassembly_rate_per_hour: 80 other · Basis not reported
Supporting report

“其整体作业成功率超过99%,每小时可拆垛80箱、码垛75箱,直接在原有仓库布局内完成自动化升级。”

Exact source · revision 2

Source owner not reported

Reported quantity · palletizing_rate_per_hour: 75 other · Basis not reported
Supporting report

“其整体作业成功率超过99%,每小时可拆垛80箱、码垛75箱,直接在原有仓库布局内完成自动化升级。”

Exact source · revision 2

Source owner not reported

Reported quantity · Reported success rate: 99 percent · Basis not reported · Differing source assertions
Supporting report

“其整体作业成功率超过99%,每小时可拆垛80箱、码垛75箱,直接在原有仓库布局内完成自动化升级。”

Exact source · revision 2

Source owner not reported

Supporting report

“其整体作业成功率超过99%,每小时可拆垛80箱、码垛75箱,直接在原有仓库布局内完成自动化升级。”

Exact source · revision 2

Source owner not reported

Reported quantity · assembly_time_per_robot: 2.5 other · Basis not reported · Differing source assertions
Supporting report

“智元机器人2.5分钟造一台“小伙伴”,还学会“无师自通”了”

Exact source · revision 2

Source owner not reported

Reported quantity · robots_assembled: 5000 robots · Basis not reported
Supporting report

“同时,启元宣布实现第5000台机器人下线,距离其开售仅过去一个月。”

Exact source · revision 2

Source owner not reported

Reported quantity · payload_capacity_single_arm: 20 other · Basis not reported
Supporting report

“精灵G2 Max单臂负载20公斤,双臂峰值负载可达50公斤,手臂重复定位精度控制在0.5毫米以内,是行业内率先实现长程、多任务作业的重载具身机器人。”

Exact source · revision 2

Source owner not reported

Reported quantity · payload_capacity_double_arms: 50 other · Basis not reported
Supporting report

“精灵G2 Max单臂负载20公斤,双臂峰值负载可达50公斤,手臂重复定位精度控制在0.5毫米以内,是行业内率先实现长程、多任务作业的重载具身机器人。”

Exact source · revision 2

Source owner not reported

Reported quantity · assembly_time_per_robot: 2.5 other · Basis not reported · Differing source assertions
Supporting report

“智元发布“造机器人”的机器人,节奏稳不瞌睡2.5分钟组装一台“同伴””

Exact source · revision 2

Source owner not reported

Reported quantity · robots_assembled_per_month: 5000 robots · Basis not reported
Supporting report

“从宣布正式发售个人机器人到第5000台机器人下线仅用时1个月、每2.5分钟下线一台机器人、日产能从此前的10余台提升至200余台……启元个人机器人稳定高效的产线背后,有一支机器人队伍。”

Exact source · revision 2

Source owner not reported

Reported quantity · single_arm_load_capacity: 20 other · Basis not reported
Supporting report

“精灵G2 Max单臂负载20kg,双臂额定负载40kg,双臂峰值负载可达50kg,手臂重复定位精度≤±0.5mm,是行业内率先实现长程、多任务作业的重载具身机器人。”

Exact source · revision 2

Source owner not reported

Reported quantity · double_arm_load_capacity: 40 other · Basis not reported
Supporting report

“精灵G2 Max单臂负载20kg,双臂额定负载40kg,双臂峰值负载可达50kg,手臂重复定位精度≤±0.5mm,是行业内率先实现长程、多任务作业的重载具身机器人。”

Exact source · revision 2

Source owner not reported

Reported quantity · double_arm_peak_load_capacity: 50 other · Basis not reported
Supporting report

“精灵G2 Max单臂负载20kg,双臂额定负载40kg,双臂峰值负载可达50kg,手臂重复定位精度≤±0.5mm,是行业内率先实现长程、多任务作业的重载具身机器人。”

Exact source · revision 2

Source owner not reported

Reported quantity · repeatability_accuracy: 0.5 other · Basis not reported · Differing source assertions
Supporting report

“精灵G2 Max单臂负载20kg,双臂额定负载40kg,双臂峰值负载可达50kg,手臂重复定位精度≤±0.5mm,是行业内率先实现长程、多任务作业的重载具身机器人。”

Exact source · revision 2

Source owner not reported

Reported quantity · boxes_disassembled_per_hour: 80 other · Basis not reported
Supporting report

“其整体作业成功率超99%,每小时可实现拆垛80箱、码垛75箱,直接在原有仓库布局内完成自动化升级。”

Exact source · revision 2

Source owner not reported

Evidence dependency changes (19)
  • 2026-10-30T19:41:14.147Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
  • 2026-10-30T19:40:57.244Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
  • 2026-10-30T19:37:04.111Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
  • 2026-10-30T19:36:49.149Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
  • 2026-10-30T19:18:57.978Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
  • 2026-10-30T19:18:45.556Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
  • 2026-10-30T19:11:46.374Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
  • 2026-10-30T19:11:36.903Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
  • 2026-10-30T19:06:53.073Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
  • 2026-10-30T19:02:50.871Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
  • 2026-10-30T18:27:21.792Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
  • 2026-10-30T18:23:12.173Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
  • 2026-10-30T17:43:53.443Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
  • 2026-10-30T17:37:36.987Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
  • 2026-10-30T17:37:36.872Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
  • 2026-10-30T17:36:21.543Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
  • 2026-10-30T17:25:57.812Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
  • 2026-10-30T17:21:51.543Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
  • 2026-10-30T17:15:58.193Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.

Developments

3 developments
  1. 2026-10-27 11:46 UTC · 2 reports
    Zhiyuan Robotics Releases G2 Max Heavy-Lift Embodied Robot, Achieving 2.5-Minute
    Chinese robotics — Deployment discovery:Zhiyuan Robotics Releases Embodied Intelligence Framework, Enabling Robot Autonomous Evolution
  2. 2026-10-26 14:30 UTC · 1 reports
    Zhuyuan Releases Heavy-Load Embodied Robot Sprite G2 Max, Achieving 7×24 Hour St
    Chinese robotics — Deployment discovery:Zhuyuan Releases Heavy-Load Embodied Robot Sprite G2 Max, Achieving 7×24 Hour Stable Operation
  3. 2026-10-25 21:20 UTC · 1 reports
    JD Logistics × Zhiyuan Robotics: Embodied Robot G2 Max Applied in Zhi Wang Wareh
    Chinese robotics — Deployment discovery:JD Logistics × Zhiyuan Robotics: Embodied Robot G2 Max Applied in Zhi Wang Warehouse

Report timeline

Follow attributed reports and material updates.

10/29
  1. Chinese robotics — Hardware and sensing
    Zhihui Peng of Zhiyuan: Whether Robots Can Work Stably Is More Critical Than Parameters

    At the 2026 AIGBOT INNOVATION DAY, Zhihui Peng, co-founder and CTO of Zhiyuan, emphasized the industry's transition from parameter-based evaluation to work-hour-based metrics. The company introduced Worken, a new standard to quantify robot productivity in real-world scenarios. Peng also discussed the challenges of physical AI, including real-time reasoning and physical task closure, and highlighted the importance of reliable hardware and data collection for deployment.

10/28
  1. Chinese robotics — Hardware and sensing
    Humanoid Robots, About to Deliver Big Results

    The article discusses the significant advancements in humanoid robotics, with Tesla launching its Optimus production line and NVIDIA promoting physical AI through its simulation platforms. It highlights the challenges in translating AI capabilities into physical tasks, the role of simulation and real-world data, and China's growing dominance in the manufacturing and supply chain for humanoid robots. The piece also touches on the potential market size and the importance of physical AI in transforming industries.

  2. Chinese robotics — Hardware and sensing
    Humanoid Robots Are About to Deliver Big Results

    The article discusses the rapid development of humanoid robots, with Tesla and NVIDIA leading the charge in production and simulation. It outlines the technical challenges of physical AI, including the need for real-world data and the role of simulation platforms like NVIDIA's Isaac. The piece also discusses the competitive landscape, with Chinese companies like Zhuyu and Zhixun leading in production, and the potential for physical AI to transform manufacturing and logistics.

  3. Chinese robotics — Deployment discoverySignal
    Zhiyuan Robotics Releases G2 Max Heavy-Lift Embodied Robot, Achieving 2.5-Minute Assembly of a 'Companion'

    Zhiyuan Robotics has launched the G2 Max, an industrial-grade heavy-lift embodied robot capable of assembling a robot in 2.5 minutes. The robot features high-precision handling, dual-arm force control, and visual fusion technology, compatible with various industrial materials without requiring extensive teaching. Additionally, the G2 Max has been deployed in convenience stores, enabling continuous learning and skill evolution in unattended environments.

10/27
  1. Chinese robotics — Deployment discoverySignal
    Zhiyuan Robotics Releases Embodied Intelligence Framework, Enabling Robot Autonomous Evolution

    Zhiyuan Robotics has launched the Sprite G2 Max, a heavy-duty embodied robot that can assemble a complete robot in 2.5 minutes, while also featuring an embodied intelligence framework that allows for autonomous learning and skill evolution in unattended environments. The robot has already been applied in e-commerce warehousing scenarios, enhancing operational efficiency and flexibility.

10/26
  1. Chinese robotics — Deployment discoverySignal
    Zhuyuan Releases Heavy-Load Embodied Robot Sprite G2 Max, Achieving 7×24 Hour Stable Operation

    Zhuyuan has introduced the safety-standard compliant heavy-load embodied robot Sprite G2 Max, targeting applications such as disassembling pallets and material transfer, supporting 7×24 hour continuous operation. The robot features dual arms with a payload capacity of 40kg and a repeat positioning accuracy of ≤±0.5mm, compatible with various industrial materials and requiring minimal teaching. Its spatial adaptability has been applied in e-commerce warehousing scenarios, enabling mixed-line operations for multiple product types.

10/25
  1. Chinese robotics — Deployment discoverySignal
    JD Logistics × Zhiyuan Robotics: Embodied Robot G2 Max Applied in Zhi Wang Warehouse

    Zhiyuan Robotics has deployed the G2 Max, an industrial-grade embodied robot, in JD Logistics' Zhi Wang warehouse for inbound handling tasks. The robot can carry up to 50kg with high precision, achieving over 99% operational success rate. This marks a real-world application of embodied intelligence, with JD Logistics emphasizing the importance of industrial scenarios for testing and refining such technologies.

Event coverage history

There is not enough continuous observation data to show a trend.

Timezone · UTC

Article dates follow your selected timezone. Briefing editions use Hong Kong time (UTC+8).