Benchmark Almanac

QAG-360K

VisTA 77.3% tops its leaderboard
Not in model cards

Models compare two satellite images, answer questions about land cover, and mark the relevant pixels. The authors score answer accuracy and mask overlap separately.

The ChangeVG team scored six general vision models on questions asking whether a change occurred. This comparison covers one of eight question types.

Made by
Ke Li and colleagues
Released
October 2024
Kind
Spatial reasoning and counting
Level
School
Scored
Checked automatically
Answers
Short answer
Tasks
All public
Measures
Capability
Language
English
In and out
Text, Images, Maps and geodata → Text, Images
Interaction
Single turn
Refresh
One fixed set
Run by
Anyone
8
Entries in its paper, 0 models we track

VisTA tops the leaderboard with 77.3%

Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. The headline is overall textual answer accuracy on QAG-360K. 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.

  1. 1 VisTA Res-101, two weight-shared visual backbones 77.3%
  2. 2 SOBA Swin-T 72%
  3. 3 ETRIS ViT-B/16 71.8%
  4. 4 LAVT Swin-B 70.9%
  5. 5 CGFormer Res-101 70.6%
  6. 6 CRIS Res-101 69.5%
  7. 7 CDVQA ViT-B/16 68.6%
  8. 8 RSVQA Res-101 67.3%

Source: The QAG-360K paper

#EntryScoreSettingDateSource
1 VisTA 77.3% Res-101, two weight-shared visual backbones Official leaderboard
2 SOBA 72% Swin-T Official leaderboard
3 ETRIS 71.8% ViT-B/16 Official leaderboard
4 LAVT 70.9% Swin-B Official leaderboard
5 CGFormer 70.6% Res-101 Official leaderboard
6 CRIS 69.5% Res-101 Official leaderboard
7 CDVQA 68.6% ViT-B/16 Official leaderboard
8 RSVQA 67.3% Res-101 Official 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

All 636 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

All 8 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 2 pass 3 notes
  • Pass Independent runs Scores from the benchmark’s paper (8 entries).
  • Pass Models scored 8 entries on its leaderboard, none a model we track.
  • Note Newest score No scores yet.
  • Note Frontier models No models scored yet.
  • Note Headroom No scores yet.
Leaderboard import 3 pass
  • Pass Importer We read the board on arxiv.org: 8 entries at the last read, none a model we track.
  • 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.

How we check