6 November 2026

AGIBOT Launches Genie Studio 2.0: From Robot Model to Workplace in Seven Days

AGIBOT Genie Studio 2.0 Agent: visuele workflow-editor met 3D-simulatie van een G2-robot

AGIBOT, the manufacturer behind the humanoid robots RaaSX offers as Benelux partner through Robots-as-a-Service, launched Genie Studio 2.0 in Shanghai on 26 October 2026. It is an integrated platform for developing, deploying and operating embodied AI applications. According to AGIBOT, initial deployment and validation in a new task environment can take as little as seven days.

Three modules, one way of working

Flywheel drives model development. Data from physical robots, simulation (Genie Sim) and open datasets (AGIBOT WORLD) comes together to train perception, vision-language-action (VLA) and reinforcement learning (RL) models, including access to AGIBOT's GO-2 foundation model. The Genie Sim Benchmark covers 42 task categories, and the gap between simulated and real-robot results is under 10%. Failure and anomaly data from the field feeds straight into the next training cycle.

Robots learning from each other's experience. Read more in our article on GO-3 Preview. Video: AGIBOT

Agent makes tasks buildable without code. In a visual drag-and-drop environment, teams combine navigation, visual localisation, decision logic and AI models into a single workflow, with connections to their own business logic and third-party systems. Every workflow is tested in simulation for paths, collisions and poses first, and can then be saved as a reusable template.

AGIBOT Genie Studio 2.0 Agent: visuele workflow-editor met 3D-simulatie van een G2-robot
Genie Studio 2.0 Agent: build and test workflows in simulation. Image: AGIBOT

Monitor gives one central view of the entire fleet: status, task progress and faults. Robots can be grouped and scheduled, with alerts, remote intervention and online debugging.

AGIBOT Genie Studio 2.0 Monitor: centraal vlootoverzicht met robotstatus en werkplekken
Genie Studio 2.0 Monitor: the whole fleet in one view. Image: AGIBOT

Proven in practice

  • CNC machine tending, Changzhou: six CNC machines and AGVs, from requirements to a validated production line in 14 days, integrated with PLCs and MES.
  • PIA Automation: three AGIBOT G2 robots assemble and pack BMS controllers with placement accuracy under 1 mm, success rates above 99.9% and cycle times of 13 seconds or less, 24/7.
  • Longcheer Technology: central management of multiple robots loading and unloading tablets, tracking operating hours, success rates and faults.
  • Guangzhou Metro: more than 60 devices across stations, with a dashboard for status, success rates and mean time between failures.

What this means for the Benelux

The three biggest barriers to robotisation are custom development, deployment time and long-term operations. Genie Studio 2.0 addresses all three. That fits how RaaSX works: with Robots-as-a-Service, monitoring, updates and preventive maintenance are part of the service, so your organisation can focus on the work itself.

It is relevant for education too. A visual workflow environment makes it more accessible for students to design, test and improve real robot tasks.

Our starting point stays the same: robots are co-workers, not replacements. They take over repetitive and heavy work, so people have time for the work where they make the difference. Earlier we wrote about the 20,000th AGIBOT robot and about Worken, AGIBOT's measure of robot productivity.

Curious what AGIBOT robots can do in your organisation? Schedule an online meeting or call +31 30 269 0012.

SHARE THIS ARTICLE
0