BURGER
Robot policies draw shapes and objects from language instructions. The authors compare how well the resulting trajectories follow the requested drawing.
- Made by
- Xiangyu Xu and colleagues
- Released
- October 2026
- Announced on X
- 7K views, 79 likes
- Kind
- Real robots
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Agent task
- Tasks
- All public
- Measures
- Capability
- Language
- English
- In and out
- Text, Images, Time series → Actions
- Interaction
- Agentic
- Environment
- A simulator, A physical robot or device
- Refresh
- One fixed set
- Run by
- Anyone
- Last model added
- April 2025
- Scores refreshed
- November 20th, 2026
π0-FAST leads with 0.56 among models released by April 2025
Best score per model from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart.
The newest model with a score here came out in April 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 16 more
Source: The BURGER paper
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | π0-FAST | 0.56 | 65 training classes, 50 demonstrations per class | Jan 2025 | Official leaderboard |
| 2 | π0.5 | 0.55 | 132 training classes, 100 demonstrations per class | Apr 2025 | Official leaderboard |
| 3 | π0 | 0.52 | 65 training classes, 100 demonstrations per class | Oct 2024 | Official leaderboard |
Scores at each effort level
Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.
Source: The BURGER paper
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| π0-FAST | 65 training classes, 50 demonstrations per class | 0.56 | |||
| π0-FAST | 132 training classes, 100 demonstrations per class | 0.55 | |||
| π0.5 | 132 training classes, 100 demonstrations per class | 0.55 | |||
| π0.5 | 30 training classes, 100 demonstrations per class | 0.54 | |||
| π0-FAST | 30 training classes, 50 demonstrations per class | 0.54 | |||
| π0 | 65 training classes, 100 demonstrations per class | 0.52 | |||
| π0.5 | 65 training classes, 100 demonstrations per class | 0.50 | |||
| π0-FAST | 132 training classes, 50 demonstrations per class | 0.48 | |||
| π0-FAST | 65 training classes, 100 demonstrations per class | 0.48 | |||
| π0-FAST | 65 training classes, 10 demonstrations per class | 0.48 | |||
| π0-FAST | 132 training classes, 10 demonstrations per class | 0.48 | |||
| π0 | 132 training classes, 100 demonstrations per class | 0.47 | |||
| π0-FAST | 30 training classes, 10 demonstrations per class | 0.47 | |||
| π0-FAST | 10 training classes, 50 demonstrations per class | 0.47 | |||
| π0 | 10 training classes, 100 demonstrations per class | 0.46 | |||
| π0.5 | 10 training classes, 100 demonstrations per class | 0.46 | |||
| π0-FAST | 30 training classes, 100 demonstrations per class | 0.46 | |||
| π0 | 132 training classes, 50 demonstrations per class | 0.46 | |||
| π0.5 | 65 training classes, 50 demonstrations per class | 0.45 | |||
| π0 | 30 training classes, 100 demonstrations per class | 0.42 | |||
| π0-FAST | 10 training classes, 10 demonstrations per class | 0.41 | |||
| π0.5 | 132 training classes, 10 demonstrations per class | 0.40 | |||
| π0.5 | 132 training classes, 50 demonstrations per class | 0.40 | |||
| π0.5 | 30 training classes, 50 demonstrations per class | 0.39 | |||
| π0 | 30 training classes, 50 demonstrations per class | 0.39 | |||
| π0 | 65 training classes, 50 demonstrations per class | 0.38 | |||
| π0-FAST | 10 training classes, 100 demonstrations per class | 0.35 | |||
| π0.5 | 65 training classes, 10 demonstrations per class | 0.34 | |||
| π0.5 | 10 training classes, 50 demonstrations per class | 0.33 | |||
| π0 | 132 training classes, 10 demonstrations per class | 0.32 | |||
| π0 | 10 training classes, 50 demonstrations per class | 0.32 | |||
| π0 | 65 training classes, 10 demonstrations per class | 0.32 | |||
| π0.5 | 30 training classes, 10 demonstrations per class | 0.31 | |||
| π0 | 30 training classes, 10 demonstrations per class | 0.23 | |||
| π0.5 | 10 training classes, 10 demonstrations per class | 0.21 | |||
| π0 | 10 training classes, 10 demonstrations per class | 0.20 |
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: The BURGER paper
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official BURGER leaderboard
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
More 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%
- SimplerEnv In use Real-to-sim replicas of Google Robot and WidowX/Bridge setups. Policies scored by success rate on pick, move, drawer and Bridge tasks. Praxis-1 80%
- RoboLab In use Zero-shot DROID Franka policies in Isaac Lab on 120 tabletop tasks at three instruction levels, 10 episodes each. Success rate. Praxis-1 46.2%
- 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 27.6%
- 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%
Data audit
7 of 8 checks pass, 1 warning. Look first at the Newest score check. The newest came 19 months ago, in April 2025, for π0.5. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 1 pass 3 notes
- Pass Independent runs Scores from the benchmark’s paper (3 entries).
- Note Models scored Only 3 models we track.
- Warning Newest score The newest came 19 months ago, in April 2025, for π0.5.
- Note Frontier models None of its models are on the capabilities index.
- Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on siyingzhao.github.io: 36 rows for models we track at the last read.
- Pass Last read November 20th, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.