Benchmark Almanac

RoboCasa GR1 Tabletop Tasks

GR00T N1 50% leads since March 2025
Fading

A simulated Fourier GR-1 humanoid with dexterous hands does 24 tabletop pick-and-place tasks, including unseen object and container pairs. Success rate per task.

Built for the GR00T N1 report with 1,000 demos per task. About 170 GitHub stars. NVIDIA still reports it for GR00T N1.7 with 20 trials per task.

Released
March 2025
Kind
Simulated robots
Level
Everyday
Scored
Checked automatically
Answers
Agent task
Tasks
All public
Last in a launch post
July 2026, dropped before it saturated
Last model added
July 2026
4
Models scored in lab cards
4
Launch posts list it, from 1 lab
1
System card reports it, from 1 lab

GR00T N1 leads with 50% among models released by April 2026

Best score per model as labs report them in their own cards. Labs pick their own settings, so these compare less cleanly than independent runs.

The newest model with a score here came out in April 2026, and nobody has reported a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 GR00T N1 50%
  2. 2 GR00T N1.5 48%
  3. 3 GR00T N1.6 47.6%
  4. 4 GR00T N1.7 44.5%

Source: Labs’ model cards

#ModelScoreSettingDateSource
1 GR00T N1 50% Mar 2025 Lab model card
2 GR00T N1.5 48% Sep 2025 Lab model card
3 GR00T N1.6 47.6% Dec 2025 Lab model card
4 GR00T N1.7 44.5% Jul 2026 Lab model card

Number one over time

Lab claims count from the day the card came out, independent runs from the model’s release.

NVIDIA
Lab claims Not tested since Jul 2026
GR00T N1 50%

Source: Labs’ model cards

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

Reported by the lab in its model card
0% 20% 40% 60% 80% 100% 2025 2026 Near the ceiling (85%) GR00T N1 0% 20% 40% 60% 80% 100% 2025 2026 Near the ceiling (85%) GR00T N1

Source: Labs’ model cards

Labs headline it less often now, last in July 2026

Launch posts from every lab that has listed it, one tick per post. Tall colored ticks include it, and short gray ones leave it out.

NVIDIA
DateLaunch postScoreAs the post writes it
Jul 7, 2026 GR00T N1.7 44.5 RoboCasa GR1 Tabletop Tasks, mean of 24 per-task success rates, 20 trials per task, finetuned N1.7 checkpoint (examples/robocasa-gr1-tabletop-tasks README)
Dec 16, 2025 GR00T N1.6 47.6 RoboCasa GR1 Tabletop Tasks, average of 24 tasks (examples/robocasa-gr1-tabletop-tasks README)
Jun 11, 2025 GR00T N1.5 47.4 Sim GR-1, 30 Demos per Task. Sim GR-1 0-shot 43.9 (data-limited post-training table)
Mar 18, 2025 GR00T N1 50 GR-1 Tabletop Tasks, 24 tasks, 100 demonstrations per task, average success rate, GR00T-N1-2B (Table 2)

One system card reports it

Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.

DateDocumentScoreAs the document writes itSection
Sep 26, 2025 GR00T N1.5 model card 48 RoboCasa GR1 tabletop tasks, average success 0.48 over 24 tasks, stored here as 48% (examples/RoboCasa README) Capabilities

More in Robotics

All 103 in Robotics
  • ERQA In use 400 multiple-choice visual questions from robot and egocentric footage on spatial, trajectory, action and state reasoning. Accuracy. Gemini Robotics ER 2 78.5%
  • RoboDojo In use Generalist manipulation policies run 42 Isaac Sim tasks and 18 remote real-robot tasks. Scored by task score and success rate. GPT-6 Astra 29%
  • RoboLab In use Zero-shot DROID Franka policies in Isaac Lab on 120 tabletop tasks at three instruction levels, 10 episodes each. Success rate. FLUX 3 Action 42.9%
  • Progress classification (Google DeepMind) In use Models assign each frame of a robot video to one of five task-progress bands from 0 to 100 percent. Scored by accuracy. Gemini Robotics ER 2 57.4%
  • Human proximity (Google DeepMind robotics) In use Models judge from robot camera views whether a human is within 1 meter, a spatial awareness check for safe robot behavior. Scored by accuracy. Gemini Robotics ER 2 93%
  • Moment finding (Google DeepMind) In use Models find the moment in a robot video when an event happens. Scored by accuracy and mean absolute time distance in seconds. Gemini Robotics ER 2 91.3%