OpenEQA
Agents answer over 1,600 open-vocabulary questions about real indoor scenes from recorded episodes or active exploration. A judge model scores answers against human answers.
- Made by
- Meta FAIR
- Released
- April 2024
- Kind
- Embodied reasoning
- Level
- Everyday
- Scored
- Graded by a model
- Answers
- Short answer
- Tasks
- All public
- Strongest model tested
-
Gemini 1.5 Pro May 2024 #173 on the capability index - Last in a launch post
- May 2024, dropped before it saturated
- Last model added
- September 2025
- Scores refreshed
- November 12th, 2026
Gemini 1.5 Pro leads with 57.9
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.
Gemini 1.0 Pro Vision leads with 44.9 across all independent runs
Best score per model across the official leaderboard, Epoch AI’s own runs and any board Epoch AI copied. Each bar and row names its source.
Gemini 1.5 Pro leads with 57.9, claims included
Best score per model from every source, including labs’ own numbers and claims Epoch AI collected from reports and papers. Labs pick their own settings, so a claim can beat an independent run of the same model.
- 1 Gemini 1.5 Pro May 2024 57.9
- 2 gemini-robotics-er-1.5-preview 55.0
Source: Labs’ model cards
- 1 Gemini 1.0 Pro Vision 15 frames 44.9
- 2 GPT-4 Jun 2023, blind 33.5
Source: Official OpenEQA leaderboard
- 1 Gemini 1.5 Pro May 2024 57.9
- 2 gemini-robotics-er-1.5-preview 55.0
- 3 Gemini 1.0 Pro Vision 15 frames 44.9
- 4 GPT-4 Jun 2023, blind 33.5
Sources: Labs’ model cards and official OpenEQA leaderboard
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Gemini 1.5 Pro May 2024 | 57.9 | May 2024 | Lab model card | |
| 2 | gemini-robotics-er-1.5-preview | 55.0 | Sep 2025 | Lab model card |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Gemini 1.0 Pro Vision | 44.9 | 15 frames | Jan 2024 | Official leaderboard |
| 2 | GPT-4 Jun 2023 | 33.5 | blind | Jun 2023 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Gemini 1.5 Pro May 2024 | 57.9 | May 2024 | Lab model card | |
| 2 | gemini-robotics-er-1.5-preview | 55.0 | Sep 2025 | Lab model card | |
| 3 | Gemini 1.0 Pro Vision | 44.9 | 15 frames | Jan 2024 | Official leaderboard |
| 4 | GPT-4 Jun 2023 | 33.5 | blind | Jun 2023 | Official leaderboard |
The benchmark's own leaderboard has 5 entries, 2 of them models we track. Switch the scores above to rank them together.
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Sources: Official OpenEQA leaderboard and labs’ model cards
No launch post has listed it since May 2024
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.
| Date | Launch post | Score | As the post writes it |
|---|---|---|---|
| May 14, 2024 | Gemini 1.5 Pro | 57.9 | OpenEQA (val), 0-shot (LLM-judge score) |
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.
| Date | Document | Score | As the document writes it | Section |
|---|---|---|---|---|
| Sep 25, 2025 | gemini-robotics-er-1.5-preview technical report | 55 | OpenEQA (Table 19, thinking on). Thinking off 50.5 | Capabilities |
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%
- 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%