Benchmark Almanac

INQUIRE-Rerank

GPT-4o Mar 2025 62.4% leads since March 2025
Not in model cards
Version Rerank

Models reorder 100 photographs supplied for each ecological query. The benchmark measures average precision among the first 50 results.

The organizers list 23 scored models plus a random baseline. The release has 200 test queries and 50 validation queries.

Made by
INQUIRE team
Released
June 2024
Kind
General visual questions
Level
University
Scored
Checked automatically
Answers
Short answer
Tasks
All public
Measures
Capability
Language
English
In and out
Text, Images → Text
Interaction
Single turn
Items from
Real work and use
Refresh
One fixed set
Run by
Anyone
Last model added
March 2025
Scores refreshed
November 20th, 2026
23
Entries on its own leaderboard, 9 models we track

GPT-4o leads with 62.4% among models released by March 2025 GPT-4o leads with 62.4% among models released by March 2025

Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. 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. This view shows only models we track, each at its best.

The newest model with a score here came out in March 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 GPT-4o Mar 2025 62.4%
  2. 2 VILA1.5-40B 54.3%
  3. 3 SigLIP SO400m-14-384 51.5%
  4. 4 GPT-4-Turbo (20240409) 48.9%
  5. 5 PaliGemma-3B-mix-448 48.9%
  6. 6 LLaVA-v1.6-34B 48.3%
  7. 7 SigLIP ViT-L/16-384 47.5%
  8. 8 VILA1.5-13B 46.3%
  9. 9 CLIP ViT-H/14-378 (DFN) 44.6%
  10. 10 InstructBLIP FLAN-T5-XXL 44.3%
  11. 11 llava-v1.6-mistral-7b 43.1%
  12. 12 LLaVA-v1.5-13B 43%
  13. 13 BLIP-2 FLAN-T5-XXL 40.5%
  14. 14 CLIP ViT-L/14 (DFN) 39.1%
  15. 15 CLIP ViT-L/14 (OpenAI) 37.8%
  16. 16 CLIP RN50x16 (OpenAI) 36.2%
  17. 17 CLIP ViT-B/16 (DFN) 33.7%
  18. 18 CLIP ViT-B/16 (OpenAI) 33.5%
  19. 19 WildCLIP-t1 31.6%
  20. 20 WildCLIP-t1t7-lwf 31.5%
  21. 21 CLIP ViT-B/32 (OpenAI) 31.3%
  22. 22 CLIP RN50 (OpenAI) 31.2%
  23. 23 BioCLIP 28.9%

Source: Official INQUIRE-Rerank leaderboard

#EntryScoreSettingDateSource
1 GPT-4o Mar 2025 62.4% Mar 2025 Official leaderboard
2 VILA1.5-40B 54.3% Apr 2024 Official leaderboard
3 SigLIP SO400m-14-384 0.9B 51.5% Official leaderboard
4 GPT-4-Turbo (20240409) 48.9% Official leaderboard
5 PaliGemma-3B-mix-448 48.9% May 2024 Official leaderboard
6 LLaVA-v1.6-34B 48.3% Jan 2024 Official leaderboard
7 SigLIP ViT-L/16-384 0.7B 47.5% Official leaderboard
8 VILA1.5-13B 46.3% Apr 2024 Official leaderboard
9 CLIP ViT-H/14-378 (DFN) 1B 44.6% Official leaderboard
10 InstructBLIP FLAN-T5-XXL 44.3% Jun 2023 Official leaderboard
11 llava-v1.6-mistral-7b 43.1% Jan 2024 Official leaderboard
12 LLaVA-v1.5-13B 43% Oct 2023 Official leaderboard
13 BLIP-2 FLAN-T5-XXL 40.5% Feb 2023 Official leaderboard
14 CLIP ViT-L/14 (DFN) 0.4B 39.1% Official leaderboard
15 CLIP ViT-L/14 (OpenAI) 0.4B 37.8% Official leaderboard
16 CLIP RN50x16 (OpenAI) 0.3B 36.2% Official leaderboard
17 CLIP ViT-B/16 (DFN) 0.2B 33.7% Official leaderboard
18 CLIP ViT-B/16 (OpenAI) 0.2B 33.5% Official leaderboard
19 WildCLIP-t1 0.2B 31.6% Official leaderboard
20 WildCLIP-t1t7-lwf 0.2B 31.5% Official leaderboard
21 CLIP ViT-B/32 (OpenAI) 0.2B 31.3% Official leaderboard
22 CLIP RN50 (OpenAI) 0.1B 31.2% Official leaderboard
23 BioCLIP 0.2B 28.9% Official leaderboard
#ModelScoreSettingDateSource
1 GPT-4o Mar 2025 62.4% Mar 2025 Official leaderboard
2 VILA1.5-40B 54.3% Apr 2024 Official leaderboard
3 PaliGemma-3B-mix-448 48.9% May 2024 Official leaderboard
4 LLaVA-v1.6-34B 48.3% Jan 2024 Official leaderboard
5 VILA1.5-13B 46.3% Apr 2024 Official leaderboard
6 InstructBLIP FLAN-T5-XXL 44.3% Jun 2023 Official leaderboard
7 llava-v1.6-mistral-7b 43.1% Jan 2024 Official leaderboard
8 LLaVA-v1.5-13B 43% Oct 2023 Official leaderboard
9 BLIP-2 FLAN-T5-XXL 40.5% Feb 2023 Official leaderboard

Rankings by part

We read 4 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as Appearance, Behavior and Context. GPT-4o leads every one of them. Arrows show how far a model moved from its overall rank among the models we track.

Appearance

9 models we track have a score here.

  1. 1 GPT-4o Mar 2025 59.7%
  2. 2 VILA1.5-40B 50.4%
  3. 3 PaliGemma-3B-mix-448 44.1%
  4. 4 LLaVA-v1.6-34B 43.7%
  5. 5 VILA1.5-13B 40.2%
  6. 6 llava-v1.6-mistral-7b ↑1 39%
  7. 7 InstructBLIP FLAN-T5-XXL ↓1 38.7%
  8. 8 LLaVA-v1.5-13B 37.7%
  9. 9 BLIP-2 FLAN-T5-XXL 32.8%

2 models have held number one on the official leaderboard since its release in June 2024

Lab claims count from the day the card came out, independent runs from the model’s release.

NVIDIA OpenAI
Own leaderboard Not tested since Mar 2025
GPT-4o 62.4%

Source: Official INQUIRE-Rerank leaderboard

Scores over time

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

Plot with other benchmarks
On the benchmark’s own leaderboard Best so far
0% 20% 40% 60% 80% 100% 2023 2024 2025 Near the ceiling (85%) GPT-4o VILA1.5-40B LLaVA-v1.6-34B InstructBLIP FLAN-T5-XXL 0% 20% 40% 60% 80% 100% 2023 2024 2025 Near the ceiling (85%) GPT-4o VILA1.5-40B

Source: Official INQUIRE-Rerank 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.

Versions

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 20 months ago, in March 2025, for GPT-4o. 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 (23 entries).
  • Pass Models scored 9 models we track, plus 14 other entries on its leaderboard.
  • Warning Newest score The newest came 20 months ago, in March 2025, for GPT-4o.
  • Note Frontier models None of its models are on the capabilities index.
  • Pass Headroom The best is 62.4%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on inquire-benchmark.github.io: 9 rows for models we track and 14 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