Flippy Can Now See in 3D: Miso Robotics Debuts Next Generati
Overview
Isaac Lab 3.0 is now generally available, offering a unified framework for robot learning with support for multiple physics, rendering, and visualization backends. Key features include Newton physics, OV libraries, and Isaac Sim integration, along with improved simulation tools and reinforcement learning compatibility.
From Isaac Lab — Releases
Event evidence and corrections
2 attributed source owners. Ownership does not establish independent confirmation. Quantities are reported separately and are never added together.
Reported quantity · simulated assets: 2000 other · Basis not reported
“SimReady Central 资源库上线 Hugging Face，提供 2000 + 工厂与机器人仿真资产；Human 1M 大规模人体数据集，包含多视角同步视频、三维手部跟踪数据，面向 Cosmos 联盟伙伴开放完整资源。”
Exact source · revision 2Source owner not reported
Reported quantity · assets: 2000 other · Basis not reported
“e AI figure out how to perform it. Announced on October 21st, 'SimReady Central' is a gateway to the assets used for that preparation. You can search for assets in OpenUSD format that contain simulation-ready information, such as robot arms, grippers, and warehouse equipment, from public datasets. The announcement highlights over 2,000 assets. OpenUSD is a framework for describing 3D scenes by combining shapes, mater”
Exact source · revision 2Source owner not reported
Environment: simulation
“SO-101 tabletop (Isaac Sim) A LeRobot EnvHub environment built natively in NVIDIA Isaac Sim 6.2: an SO-101 arm in the turquoise of the LeRobot kits, four painted wooden 3 cm cubes and a green open box on a table in a simple room, simulated by PhysX and rendered with RTX. The layout, cameras, observations, actions and LEROBOT_SIM match the MuJoCo version, lerobot/so101-tabletop-mujoco , so the same policies and comman”
Exact source · revision 6huggingface
Reported quantity · observation pixels: 480 other · Basis not reported · Differing source assertions
“ds run against both. Observation pixels : front (scene camera) and wrist (gripper camera), 480×640 RGB; agent_pos : 6 joint positions Action 6 absolute joint position targets: shoulder_pan , shoulder_lift , elbow_flex , wrist_flex , wrist_roll in degrees, gripper 0-100 (the units of LeRobot's so101_follower with use_degrees=True ) Control rate 30 fps (8 PhysX steps per control step) Task "Put all the cubes in the gre”
Exact source · revision 6huggingface
Reported quantity · observation pixels: 640 other · Basis not reported · Differing source assertions
“ds run against both. Observation pixels : front (scene camera) and wrist (gripper camera), 480×640 RGB; agent_pos : 6 joint positions Action 6 absolute joint position targets: shoulder_pan , shoulder_lift , elbow_flex , wrist_flex , wrist_roll in degrees, gripper 0-100 (the units of LeRobot's so101_follower with use_degrees=True ) Control rate 30 fps (8 PhysX steps per control step) Task "Put all the cubes in the gre”
Exact source · revision 6huggingface
Reported quantity · control rate: 30 other · Basis not reported
“ds run against both. Observation pixels : front (scene camera) and wrist (gripper camera), 480×640 RGB; agent_pos : 6 joint positions Action 6 absolute joint position targets: shoulder_pan , shoulder_lift , elbow_flex , wrist_flex , wrist_roll in degrees, gripper 0-100 (the units of LeRobot's so101_follower with use_degrees=True ) Control rate 30 fps (8 PhysX steps per control step) Task "Put all the cubes in the gre”
Exact source · revision 6huggingface
Reported quantity · episode length: 3600 other · Basis not reported
“en box." Reward the fraction of cubes in the box at each step (0, 0.25, ... 1); info["cubes_in_box"] gives the count Episode end success (terminated) when all four cubes are in the box; truncated after 3600 steps (2 minutes), set with episode_length Reset seeded cube positions and yaws, joint noise Isaac Sim starts once per process, in make_env : about 20 s, plus a few minutes of shader compilation the first time on”
Exact source · revision 6huggingface
Reported quantity · step time: 80 other · Basis not reported
“a machine or in a container. It holds one world, so make_env takes n_envs=1 . A step takes about 80 ms on an RTX 3080 Laptop GPU, so lockstep evaluation is exact, while realtime runs slower than 30 fps. The arm asset is bundled, so the environment needs no network access at runtime. With the LeRobot simulator server From a LeRobot checkout, start the server. The image installs packages.txt (system libraries) and requ”
Exact source · revision 6huggingface
Reported quantity · fps: 30 other · Basis not reported
“irements.txt (Isaac Sim 6.2 from PyPI and NVIDIA's index, Python 3.12); it needs an NVIDIA RTX GPU and the NVIDIA Container Toolkit: uv run scripts/run_sim.py --envhub lerobot/so101-tabletop-isaac Then run a policy against it, for example: lerobot-rollout --robot.type=remote --robot.endpoint=tcp/127.0.0.1:7448 \ --robot.deployment=envhub --robot.task=so101_tabletop \ --policy.path=<your SO-101 checkpoint> --fps=30 Ob”
Exact source · revision 6huggingface
Reported quantity · number of cubes: 4 other · Basis not reported
“en box." Reward the fraction of cubes in the box at each step (0, 0.25, ... 1); info["cubes_in_box"] gives the count Episode end success (terminated) when all four cubes are in the box; truncated after 3600 steps (2 minutes), set with episode_length Reset seeded cube positions and yaws, joint noise Isaac Sim starts once per process, in make_env : about 20 s, plus a few minutes of shader compilation the first time on”
Exact source · revision 6huggingface
Reported quantity · reward fraction: 0 other · Basis not reported
“en box." Reward the fraction of cubes in the box at each step (0, 0.25, ... 1); info["cubes_in_box"] gives the count Episode end success (terminated) when all four cubes are in the box; truncated after 3600 steps (2 minutes), set with episode_length Reset seeded cube positions and yaws, joint noise Isaac Sim starts once per process, in make_env : about 20 s, plus a few minutes of shader compilation the first time on”
Exact source · revision 6huggingface
Reported quantity · height: 1.4 other · Basis not reported
“Agile ONE SN 本体参数：身高 1.4 米，整机 40 公斤，躯体拥有 31 个自由度，单臂最大负载 3 公斤;配备两套 WUJI Hand 2 五指灵巧手，单只手具备 20 自由度，支持 1kHz 力控与触觉感知。”
Exact source · revision 2Source owner not reported
Reported quantity · weight: 40 other · Basis not reported
“Agile ONE SN 本体参数：身高 1.4 米，整机 40 公斤，躯体拥有 31 个自由度，单臂最大负载 3 公斤;配备两套 WUJI Hand 2 五指灵巧手，单只手具备 20 自由度，支持 1kHz 力控与触觉感知。”
Exact source · revision 2Source owner not reported
Reported quantity · degrees of freedom: 31 other · Basis not reported · Differing source assertions
“Agile ONE SN 本体参数：身高 1.4 米，整机 40 公斤，躯体拥有 31 个自由度，单臂最大负载 3 公斤;配备两套 WUJI Hand 2 五指灵巧手，单只手具备 20 自由度，支持 1kHz 力控与触觉感知。”
Exact source · revision 2Source owner not reported
Reported quantity · maximum load per arm: 3 other · Basis not reported
“Agile ONE SN 本体参数：身高 1.4 米，整机 40 公斤，躯体拥有 31 个自由度，单臂最大负载 3 公斤;配备两套 WUJI Hand 2 五指灵巧手，单只手具备 20 自由度，支持 1kHz 力控与触觉感知。”
Exact source · revision 2Source owner not reported
Reported quantity · degrees of freedom per hand: 20 other · Basis not reported
“Agile ONE SN 本体参数：身高 1.4 米，整机 40 公斤，躯体拥有 31 个自由度，单臂最大负载 3 公斤;配备两套 WUJI Hand 2 五指灵巧手，单只手具备 20 自由度，支持 1kHz 力控与触觉感知。”
Exact source · revision 2Source owner not reported
Reported quantity · force control frequency: 1 other · Basis not reported
“Agile ONE SN 本体参数：身高 1.4 米，整机 40 公斤，躯体拥有 31 个自由度，单臂最大负载 3 公斤;配备两套 WUJI Hand 2 五指灵巧手，单只手具备 20 自由度，支持 1kHz 力控与触觉感知。”
Exact source · revision 2Source owner not reported
Reported quantity · compute power: 2070 other · Basis not reported
“机载 Jetson Thor 可输出 2070 FP4 TF 算力，搭配 128GB 统一内存，满足本体实时推理与运动控制;整机集成头部、腕部相机、激光雷达、IMU 多模态感知设备。”
Exact source · revision 2Source owner not reported
Reported quantity · bytes: 403375789 other · Basis not reported · Differing source assertions
“e demo/model.onnx Single-robot pipeline, including observation processing and action conversion 403,375,789 bytes racer/network.onnx Six-robot batched network; requires G1 Racer's separate observation processing 403,036,698 bytes SHA256 digests: demo/model.onnx : 7180e82cae7ec04ff7766c262d430f57573d93678e025d8a71916885072a0cd4 racer/network.onnx : 130a22718a34c0da488e2157553b40efedfefcf6ca6cb6b97cfdd67f8dfbeaf3 The R”
Exact source · revision 7nvidia
Reported quantity · bytes: 403036698 other · Basis not reported · Differing source assertions
“e demo/model.onnx Single-robot pipeline, including observation processing and action conversion 403,375,789 bytes racer/network.onnx Six-robot batched network; requires G1 Racer's separate observation processing 403,036,698 bytes SHA256 digests: demo/model.onnx : 7180e82cae7ec04ff7766c262d430f57573d93678e025d8a71916885072a0cd4 racer/network.onnx : 130a22718a34c0da488e2157553b40efedfefcf6ca6cb6b97cfdd67f8dfbeaf3 The R”
Exact source · revision 7nvidia
Reported quantity · simulation-ready robotics assets: 2000 other · Basis not reported
“ing, creating simulation environments and deploying robot applications. These tools support development workflows from data preparation through deployment. SimReady Central Library: Provides more than 2,000 simulation-ready robotics and factory assets. Human 1M Dataset: Includes human activities, synchronized three-view video and 3D hand tracking. Both the SimReady Central Library and the Human 1M Dataset are availab”
Exact source · revision 2Source owner not reported
Reported quantity · kilograms: 40 other · Basis not reported · Differing source assertions
“gile ONE SN features: Agile ONE S humanoid body: Measures 1.40 meters (4.59 feet) tall, weighs 40 kilograms (88 pounds), provides 31 degrees of freedom and supports a 3-kilogram payload per arm. Dual WUJI Hand 2 hands: Each five-finger hand provides 20 degrees of freedom, 1 kHz force control and tactile sensing for manipulation tasks. NVIDIA Jetson Thor onboard computing: Provides 2,070 FP4 teraflops of compute and 1”
Exact source · revision 2Source owner not reported
Reported quantity · degrees of freedom: 31 other · Basis not reported · Differing source assertions
“gile ONE SN features: Agile ONE S humanoid body: Measures 1.40 meters (4.59 feet) tall, weighs 40 kilograms (88 pounds), provides 31 degrees of freedom and supports a 3-kilogram payload per arm. Dual WUJI Hand 2 hands: Each five-finger hand provides 20 degrees of freedom, 1 kHz force control and tactile sensing for manipulation tasks. NVIDIA Jetson Thor onboard computing: Provides 2,070 FP4 teraflops of compute and 1”
Exact source · revision 2Source owner not reported
Reported quantity · kilograms: 3 other · Basis not reported · Differing source assertions
“gile ONE SN features: Agile ONE S humanoid body: Measures 1.40 meters (4.59 feet) tall, weighs 40 kilograms (88 pounds), provides 31 degrees of freedom and supports a 3-kilogram payload per arm. Dual WUJI Hand 2 hands: Each five-finger hand provides 20 degrees of freedom, 1 kHz force control and tactile sensing for manipulation tasks. NVIDIA Jetson Thor onboard computing: Provides 2,070 FP4 teraflops of compute and 1”
Exact source · revision 2Source owner not reported
Reported quantity · kilograms: 88 other · Basis not reported · Differing source assertions
“gile ONE SN features: Agile ONE S humanoid body: Measures 1.40 meters (4.59 feet) tall, weighs 40 kilograms (88 pounds), provides 31 degrees of freedom and supports a 3-kilogram payload per arm. Dual WUJI Hand 2 hands: Each five-finger hand provides 20 degrees of freedom, 1 kHz force control and tactile sensing for manipulation tasks. NVIDIA Jetson Thor onboard computing: Provides 2,070 FP4 teraflops of compute and 1”
Exact source · revision 2Source owner not reported
Reported quantity · feet: 4.59 other · Basis not reported
“gile ONE SN features: Agile ONE S humanoid body: Measures 1.40 meters (4.59 feet) tall, weighs 40 kilograms (88 pounds), provides 31 degrees of freedom and supports a 3-kilogram payload per arm. Dual WUJI Hand 2 hands: Each five-finger hand provides 20 degrees of freedom, 1 kHz force control and tactile sensing for manipulation tasks. NVIDIA Jetson Thor onboard computing: Provides 2,070 FP4 teraflops of compute and 1”
Exact source · revision 2Source owner not reported
Evidence dependency changes (28)
- 2026-10-30T22:52:40.246Z · evidence updated · source revision 6. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-30T22:52:25.638Z · evidence updated · source revision 6. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-30T22:22:27.330Z · evidence updated · source revision 1. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-30T22:22:11.261Z · evidence updated · source revision 1. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-30T20:41:59.788Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-30T20:41:38.186Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-30T20:29:08.704Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-30T20:28:46.030Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-30T19:37:46.691Z · evidence updated · source revision 7. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-30T19:37:25.838Z · evidence updated · source revision 7. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-30T19:36:03.174Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-30T19:35:48.115Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-30T19:30:49.758Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-30T19:30:30.574Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-30T18:53:49.245Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-30T18:17:27.210Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-30T17:41:30.434Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-30T17:38:00.653Z · evidence updated · source revision 7. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-30T17:37:36.384Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-30T17:35:54.251Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-30T17:29:10.117Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-29T22:53:50.507Z · evidence updated · source revision 7. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-29T17:18:42.550Z · evidence updated · source revision 6. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-29T03:57:31.271Z · evidence updated · source revision 5. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-29T02:50:42.362Z · evidence updated · source revision 6. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-28T13:28:59.760Z · evidence updated · source revision 4. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-28T03:26:40.045Z · evidence updated · source revision 5. Evidence extraction was updated; current source attributions are shown above.
- 2026-10-23T18:19:59.286Z · evidence updated · source revision 2. Evidence extraction was updated; current source attributions are shown above.
Report timeline
Follow attributed reports and material updates.
- Robotics — Robot learning news discoverySignalSimReady Central: A Gateway to Simulation-Ready 3D Assets for Robot AI Training
SimReady Central is a public repository offering over 2,000 simulation-ready 3D assets in OpenUSD format, designed for robotics research and AI training. It includes robot arms, grippers, and warehouse equipment with detailed physics properties, enabling more realistic and task-specific testing environments for robot learning.
- Chinese robotics — Deployment discoverySignalNVIDIA upgrades Isaac robot platform, collaborates with Agile Robots to launch humanoid reference platform Agile ONE SN
NVIDIA has upgraded its Isaac open robot development platform, introducing a humanoid robot hardware reference platform, new runtime, simulation toolchain, public datasets, and open workflows. The Agile ONE SN, developed in collaboration with Agile Robots, is designed for academic and research use, featuring 31 degrees of freedom, two WUJI Hand 2 hands, and Jetson Thor for edge computing. The platform aims to streamline the transition from simulation to real-world deployment.
- NVIDIA — Robotics model repository updatesSignalNVIDIA Releases Metatron-v1 Model for G1 Humanoid Control and Recovery
NVIDIA has released the Metatron-v1 model, which includes ONNX inference exports for a unified scene/recovery policy used in their G1 humanoid robot demos. The model supports interactive control, locomotion, and recovery evaluation, with specific runtime interfaces and dependencies. The exports are not interchangeable and require matching observation and control interfaces for use.
- Chinese robotics — Deployment discoverySignalNVIDIA upgrades Isaac robot platform, collaborates with Agile Robots to launch humanoid reference platform Agile ONE SN
NVIDIA has upgraded its Isaac open robot development platform, introducing a humanoid robot hardware reference platform, new runtime, simulation toolchain, public datasets, and open workflows. Agile Robots has released the Agile ONE SN humanoid reference platform, which is deeply integrated with NVIDIA's technology stack and includes Isaac GR00T, Jetson Thor, and multi-sensor capabilities. The platform aims to streamline the development process from data generation to real-world deployment.
- Robotics & Automation NewsComau Integrates Nvidia Technology to Accelerate Development of AI-Powered Robots
Comau is combining its robotics expertise with Nvidia's physical AI simulation libraries and accelerated computing to develop more intelligent and autonomous robotic systems. The collaboration includes integrating Nvidia Jetson edge compute into industrial robot controllers, using Nvidia Isaac Sim for robot portfolio integration, and optimizing vision-based features and control strategies for intralogistics products. The partnership also involves contributing SimReady OpenUSD assets to Hugging Face and training AI models for Comau's MI.RA vision system using reinforcement learning.
- Robotics — Robot learning news discoverySignalNVIDIA Expands Isaac Tools for Humanoid Robot Development
NVIDIA has expanded its Isaac open robotics development platform with a humanoid reference platform, new datasets, and runtime frameworks to support robot learning workflows from simulation through deployment. The Agile ONE SN platform integrates simulation, training, and deployment tools, offering a full-stack solution for humanoid robotics research and development.
- Industrial robotics — News discoveryComau Advances Physical AI for Industrial Robotics with NVIDIA
Comau is enhancing its industrial robotics solutions with NVIDIA's physical AI simulation and accelerated computing technologies, enabling more intelligent and autonomous robotic systems. The collaboration includes integrating NVIDIA Jetson edge compute into controllers, using Isaac Sim for simulation and testing, and training AI models for vision-based applications like palletising and depalletising.
- Chinese robotics — Hardware and sensingTesla and NVIDIA Advance Humanoid Robotics Platforms
Tesla begins Optimus production, NVIDIA upgrades Isaac platform for 2027 humanoid robotics.
- Robotics — Robot learning news discoverySignalReference Workflows for Humanoid Robot Development
NVIDIA has expanded its Isaac robotics platform with reference workflows, simulation assets, and training datasets to aid in the development of humanoid robots and industrial automation systems. These resources provide reusable software frameworks for training and testing robotic behaviors, reducing the need for custom simulation environments and AI pipelines.
- Isaac Lab — ReleasesSignalIsaac Lab 3.0 Released: Multi-Backend Architecture for Robot Learning
Isaac Lab 3.0 is now generally available, offering a unified framework for robot learning with support for multiple physics, rendering, and visualization backends. Key features include Newton physics, OV libraries, and Isaac Sim integration, along with improved simulation tools and reinforcement learning compatibility.
- LeRobot — Model repository updatesSignalLeRobot SO-101 Tabletop Environment in Isaac Sim 6.2
LeRobot releases an Isaac Sim 6.2 environment for the SO-101 robotic arm, featuring a simulated tabletop setup with wooden cubes and a green box. The environment supports RGB observations from front and wrist cameras, and is compatible with policies trained in MuJoCo. It can be run locally with an NVIDIA RTX GPU and the Isaac Sim simulator.
Related events
Timezone · UTC
Article dates follow your selected timezone. Briefing editions use Hong Kong time (UTC+8).