LAS VEGAS, Jan. 6, 2026 /PRNewswire/ -- AGIBOT, a leading robotics company specializing in embodied intelligence, today introduced Genie Sim 3.0 at CES 2026, a next-generation robot simulation platform integrated with NVIDIA Isaac Sim, an open-source, reference framework built on NVIDIA Omniverse. Genie Sim 3.0 delivers a unified, open simulation workflow that brings together digital asset generation, scene generalization, data collection, automated evaluation, and physics-based simulation in a single toolchain. Its core module, Genie Sim Benchmark, is designed as a standardized evaluation system to establish the most accurate and authoritative benchmarks for embodied intelligence.

- AGIBOT states that Genie Sim 3.0 provides the largest open-source simulation dataset in embodied AI, with more than 10,000 hours of synthetic dataset including real-world robot operation scenarios.

- Genie Sim 3.0 integrates 3D reconstruction with visual generation to create a high-fidelity simulation environment.

- It pioneers LLM-driven technology to generate vast simulation scenes and evaluation metrics in minutes.

- The evaluation system covers 200+ tasks across 100,000+ scenarios to build a comprehensive capability profile for models.

The platform will significantly accelerate model development, reduce reliance on physical hardware, and empower innovation in embodied intelligence.

From Real-World Capture to Language-Driven Scene Generation

As the industry's first open-source simulation platform to deeply integrate 3D reconstruction, visual generation, and physics engine, Genie Sim 3.0 delivers a real-time, high-fidelity, and high-precision simulation environment, addressing both the visual constraints of traditional rendering and the physical inaccuracies of video-world model. Real-world environments are captured via Skyland Innovation's MetaCam handheld 3D laser scanner, combining high-resolution RGB images, 360° LiDAR point clouds, and centimeter-level RTK positioning. Interactable objects can be converted into simulation-ready assets from a single 60-second orbital video, significantly accelerating scene construction.

A defining feature of Genie Sim 3.0 is its LLM-driven natural-language scene generation. Users can describe environments conversationally, and the system automatically produces structured scenes, visual previews, and thousands of semantic variations without manual logic coding. Vision-language models further refine and tune these scenes to meet specification-level needs, enabling rapid adaptation and strong model generalization.

A Full-Stack Pipeline for Training and Evaluation

AGIBOT has also released over 10,000 hours of open-source synthetic data covering more than 200 tasks with multi-sensor modalities such as RGB-D, stereo vision, and whole-body kinematics. An intelligent data-collection toolkit supports both low-latency teleoperation and automated task programming, paired with auto-annotation and a recovery mechanism that resumes collection after task failures--reducing the cost and time of dataset production.

For evaluation, Genie Sim 3.0 provides 100,000+ simulation scenarios, moving beyond single-metric benchmarks to construct full-spectrum capability profiles for embodied models. Leveraging LLMs, the platform auto-generates executable evaluation workflows across diverse semantic, spatial-reasoning, and manipulation dimensions, clearly outlining model strengths, limitations, and optimization pathways.

Bringing Real Industrial Environments into Simulation

Importantly, Genie Sim 3.0 is also the first simulation platform to integrate real industrial-scene datasets into both training and evaluation pipelines. It reconstructs 1:1 digital twins of logistics centers, power-inspection environments, material-handling sites, and industrial production lines, enabling end-to-end model development and testing without deploying physical robots on-site. This approach significantly shortens algorithm-validation cycles and reduces dependence on hardware infrastructure.

With its fully open-source assets, data, and codebase, Genie Sim 3.0 aims to accelerate embodied AI development while providing a scalable, authoritative benchmark ecosystem for both research institutions and industrial robotics teams.

All simulation assets, datasets, and automated evaluation source code are fully open source. Developers and researchers can access AGIBOT Genie Sim on GitHub: https://github.com/AgibotTech/genie_sim

For more information AGIBOT, please visit AGIBOT at AGIBOT.com and follow AGIBOT on:

https://www.facebook.com/AGIBOT.zhiyuan
https://x.com/AGIBOT_zhiyuan
https://www.instagram.com/AGIBOT_
https://www.youtube.com/@AGIBOT-robot

About AGIBOT

AGIBOT is dedicated to driving innovation through the integration of AI and robotics, creating world-leading general-purpose embodied robot products and an application ecosystem. Built on the foundation of the robotic body and powered by the fusion of interaction, manipulation, and locomotion intelligence -- "1 Robotic Body + 3 Intelligence" -- AGIBOT is a leading robotics company in the industry to deliver a complete product portfolio and deploy across all major application scenarios.

Photo -
Photo -
Photo -

View original content:https://www.prnewswire.co.uk/news-releases/agibot-unveils-genie-sim-3-0-at-ces-2026--the-first-open-source-simulation-platform-built-on-real-world-robot-operations-302653683.html

Images

Tags

More Like This

AGIBOT Unveils New Generation of Embodied AI Robots and Models, Accelerating Real-World Deployment of Physical AI

AGIBOT Makes Its U.S. Market Debut at CES 2026 with Its Full Humanoid Robot Portfolio

AGIBOT Declares 2026 "Deployment Year One" at APC 2026, Accelerating the Era of Embodied AI Productivity

AGIBOT Makes Debut at Fortune Event with Full-Size Humanoid Robot AGIBOT A2 as Special Guest

AGIBOT Reaches 10,000 Units as Real-World Demand for Robots Accelerates

AgiBot Achieves First Real-World Deployment of Reinforcement Learning in Industrial Robotics

Omdia Ranks AGIBOT No.1 Worldwide in Humanoid Robot Shipments in 2025

AgiBot Robotics Debuted at IROS 2025, the AgiBot World Challenge Concluded Successfully