VRSBench
Models caption aerial images, locate objects described in words, and answer visual questions. The held-out files contain 9,350 captions, 16,159 object descriptions, and 37,409 questions.
TerraAgent scored 80.2% on the visual questions under the authors’ grading protocol. Captioning and object localization have separate scores.
- Released
- June 2024
- Kind
- General visual questions
- Level
- Everyday
- Scored
- Graded by a model
- Answers
- Open-ended
- Tasks
- All public
- Measures
- Capability
- Subject
- Social and economic geography
- Language
- English
- In and out
- Images, Maps and geodata, Text → Text
- Interaction
- Single turn
- Items from
- Real work and use
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Vision
- Last model added
- November 2023
- Scores refreshed
- November 20th, 2026
Mini-Gemini tops the leaderboard with 77.8% GPT-4V leads with 65.6%
Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. The authors give no combined score. The headline is accuracy over all visual questions in Table 3. 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 GPT-4V, at 65.6%. This view shows only models we track, each at its best.
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Mini-Gemini | 77.8% | Official leaderboard | ||
| 2 | LLaVA-1.5 | 76.4% | Official leaderboard | ||
| 3 | GeoChat | 76% | Official leaderboard | ||
| 4 | GPT-4V | 65.6% | Nov 2023 | Official leaderboard | |
| 5 | MiniGPT-v2 | 38.2% | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-4V | 65.6% | Nov 2023 | 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 VRSBench 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 Vision
- MMMU In use 11.5K college-level multimodal questions across 30 subjects. Multiple-choice and open accuracy.
Qwen 3.6 Plus 86%
- MMMU-Pro In use Harder MMMU: college-level multimodal questions with more options and vision-only inputs to remove text shortcuts. GPT-5.6 Sol 84.6%
- CharXiv In use Chart understanding on real arXiv figures. Reasoning questions require synthesizing information from complex charts. Claude Fable 5 93.5%
- MathVista In use Math reasoning in visual contexts (charts, geometry, puzzles) from 31 datasets. Answer accuracy.
Seed2.1 Pro 90.7%
- MATH-Vision (MathVision) In use 3,040 competition math problems with visual context. Answer accuracy.
Qwen 3.8 Max 97.7%
- MMStar In use Models answer 1,500 human-curated image questions that cannot be solved without the image, across 6 core capabilities. Scored by accuracy.
Qwen 3.8 Max 85.9%
Data audit
9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 3 years ago, in November 2023, for GPT-4V. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
- Pass Independent runs Scores from the official leaderboard (5 entries).
- Pass Models scored 1 model we track, plus 4 other entries on its leaderboard.
- Warning Newest score The newest came 3 years ago, in November 2023, for GPT-4V.
- Note Frontier models None of its models are on the capabilities index.
- Pass Headroom The best is 65.6%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on github.com: 1 row for models we track and 4 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.