OceanGym
Agents interpret underwater images and sonar, then navigate simulated search and inspection tasks. The simulator scores how closely they reach target locations.
The authors evaluated seven multimodal models. Their repository had 155 stars on November 20, 2026. The four perception files contain 170 scored items.
- Made by
- Zhejiang University and collaborators
- Released
- September 2025
- Kind
- Simulated robots
- Level
- University
- Scored
- Checked automatically
- Answers
- Agent task
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Agentic
- Environment
- OceanGym
- Items from
- Generated
- Refresh
- One fixed set
- Run by
- Anyone
- Last model added
- August 2025
- Scores refreshed
- November 20th, 2026
GPT-4o-mini (snapshot unspecified) tops the leaderboard with 17.2 GLM-4.5V leads with 14.6
Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. The site reports separate environment averages, so the headline is the plain mean of its 12 decision tasks. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. The best model we track is GLM-4.5V, at 14.6. This view shows only models we track, each at its best.
- 1 GPT-4o-mini (snapshot unspecified) 17.2
- 2 GLM-4.5V 14.6
- 3 Gemini-2.5 11.7
- 4
Qwen2.5-VL-7B 9.97
Source: Official OceanGym leaderboard
- 1 GLM-4.5V 14.6
- 2
Qwen2.5-VL-7B 9.97
Source: Official OceanGym leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-4o-mini (snapshot unspecified) | 17.2 | Official leaderboard | ||
| 2 | GLM-4.5V | 14.6 | Aug 2025 | Official leaderboard | |
| 3 | Gemini-2.5 | 11.7 | Official leaderboard | ||
| 4 | | 9.97 | Jan 2025 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GLM-4.5V | 14.6 | Aug 2025 | Official leaderboard | |
| 2 | | 9.97 | Jan 2025 | Official leaderboard |
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official OceanGym 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 15 months ago, in August 2025, for GLM-4.5V. 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 official leaderboard (4 entries).
- Note Models scored Only 2 models we track, plus 2 other entries on its leaderboard.
- Warning Newest score The newest came 15 months ago, in August 2025, for GLM-4.5V.
- 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 oceangpt.github.io: 2 rows for models we track and 2 other entries 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.