Benchmark Almanac

CoralVQA

InternVL2.5(FT) 76.4% tops its leaderboard Claude 3.5 Haiku 41.8% leads the models we track
Not in model cards

Models answer questions about coral identity, growth, and health in reef photographs. GPT-4o checks answers against references across 16 question types.

The StochasT team tested models on 3,896 questions from 487 images using Gemini 3 Flash as judge. The original test contains 27,984 questions.

Released
May 2025
Kind
General visual questions
Level
University
Scored
Graded by a model
Answers
Short answer
Tasks
All public
Measures
Capability
Language
English
In and out
Images, Text → Text
Interaction
Single turn
Refresh
One fixed set
Run by
Anyone
Strongest model tested
Claude 3.5 Haiku #175 on the capability index
Last model added
October 2024
Scores refreshed
November 20th, 2026
6
Entries in its paper, 1 model we track

InternVL2.5(FT) tops the leaderboard with 76.4% Claude 3.5 Haiku leads with 41.8%

Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. The paper reports two group averages, so this is the plain mean of the basic visual and ecological and health accuracies. 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 Claude 3.5 Haiku, at 41.8%. This view shows only models we track, each at its best.

  1. 1 InternVL2.5(FT) 76.4%
  2. 2 Qwen2.5VL(FT) 65.5%
  3. 3 BLIP3(FT) 65.3%
  4. 4 GPT-4o (snapshot unspecified) 49.2%
  5. 5 Claude 3.5 Haiku 41.8%
  6. 6 Mini-Gemini(FT) 23.4%

Source: The CoralVQA paper

Source: The CoralVQA paper

#EntryScoreSettingDateSource
1 InternVL2.5(FT) 76.4% Official leaderboard
2 Qwen2.5VL(FT) 65.5% Official leaderboard
3 BLIP3(FT) 65.3% Official leaderboard
4 GPT-4o (snapshot unspecified) 49.2% Official leaderboard
5 Claude 3.5 Haiku 41.8% Oct 2024 Official leaderboard
6 Mini-Gemini(FT) 23.4% Official leaderboard
#ModelScoreSettingDateSource
1 Claude 3.5 Haiku 41.8% Oct 2024 Official leaderboard

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

In the benchmark’s paper
0% 20% 40% 60% 80% 100% 2025 Near the ceiling (85%) Claude 3.5 Haiku 0% 20% 40% 60% 80% 100% 2025 Near the ceiling (85%) Claude 3.5 Haiku

Source: Official CoralVQA 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

9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 2 years ago, in October 2024, for Claude 3.5 Haiku. 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 benchmark’s paper (6 entries).
  • Pass Models scored 1 model we track, plus 5 other entries on its leaderboard.
  • Warning Newest score The newest came 2 years ago, in October 2024, for Claude 3.5 Haiku.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Claude 3.5 Haiku (#175).
  • Pass Headroom The best is 41.8%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on arxiv.org: 1 row for models we track and 5 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.

How we check