Holo Engine on RoboDojo

Benchmark results for a dual-arm ARX X5 in our CPU-based MuJoCo replica. HoloLang selects the robot’s actions, with Holo Reason supporting recognition.

Recorded run · Sorting rubber ducks by sizeWatch RoboDojo demoView task details

Interim results ·

Leaderboard comparison

The overall score weights five benchmark dimensions equally. The chart compares our local MuJoCo run with selected official Isaac Sim entries.

This score reflects the current Holo policy and the layouts tested so far. We’ll report new results as we refine the policy and run further evaluations.

Official RoboDojo simulation leaderboard

How Holo learns and acts

We prepare examples and visual references from RoboDojo data. During each attempt, Holo Engine uses those references and fresh measurements to select the next skill and, where the skill calls for it, fit a rule for the movement.

Before evaluation

RoboDojo inputs

Expert demonstrations, public layouts and 3D object models.

Extract and render

Measure grasp and release examples. Render object views and silhouettes.

Task knowledge

Save motion examples, task parameters and visual references.

Task knowledge feeds perception and Holo Engine during the attempt.

During each attempt

MuJoCo scene

RGB views, arm poses, gripper opening and the task instruction.

Python perception

Extract visual features and estimate object geometry from camera views.

Holo Engine
  • Holo Reason identifies objects from reference examples.
  • HoloLang tracks progress and selects the next skill.
  • Function Tree fits and evaluates movement rules for selected skills.
Motion skills

Build arm and gripper trajectories for the selected action.

Motion changes the scene. Fresh camera views and grip checks feed the next decision.
Local evaluation hardware
MacBook Pro
Intel Core i9-9880H · CPU only
HoloLang decision time
431 ms on average
Holo Engine peak RAM
88.1 MB

Peak RAM. We measured the peak resident memory of the Holo Engine process across 7 sampled tasks on CPU, starting a fresh process for each task.

HoloLang. Timing covers 2,397 logged decisions across 52 configurations, including Function Tree when called within a decision.

Function Tree. Fitting and evaluating the block stacking rule took 23 ms on average across 30 calls.

Holo Reason. In separate measurements, it took 1.56 s for two digits, 5.19 s for five objects and 21.66 s to match a text description against the full reference library. Each figure averages 3 requests.

All 42 tasks

Select a task to see the recorded run. Where a task has a randomized variant, its row averages the standard and randomized configurations.

Recorded runs

Watch the best recorded Holo attempt for each tested configuration. Select “Watch RoboDojo demo” to play the official reference video. The reference streams online, and its layout may differ from the Holo run.

Method and limits

Every robot. A measurable outcome.

Tell us what your robot needs to do. Let’s discuss a pilot.

Request a pilot

Tell us what your robot needs to do and what a successful pilot would demonstrate.

Tell us about your task. We’ll discuss the next step.

You can also reach us at hi@hiholo.ai.

RoboDojo demo

Open official task page ↗