Benchmark Almanac

PISA-Bench

GPT-4o (snapshot unspecified) 67.8% tops its leaderboard Qwen2.5-VL-72B 63.3% leads the models we track
Not in model cards

Models answer illustrated PISA questions in six languages. Parallel questions let researchers compare language effects on reading, science and mathematical reasoning.

Released
October 2025
Kind
General visual questions
Level
School
Scored
Checked automatically
Answers
Multiple choice
Tasks
All public
Measures
Capability
Language
Many languages
In and out
Text, Images → Text
Interaction
Single turn
Items from
Exams and contests
Refresh
One fixed set
Run by
Anyone
Strongest model tested
Gemma 3 27B #165 on the capability index
Last model added
October 2025
Scores refreshed
November 20th, 2026
19
Entries on its own leaderboard, 15 models we track

GPT-4o (snapshot unspecified) tops the leaderboard with 67.8% Qwen2.5-VL-72B leads with 63.3% among models released by October 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. The best model we track is Qwen2.5-VL-72B, at 63.3%. This view shows only models we track, each at its best.

The newest model with a score here came out in October 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 (snapshot unspecified) 67.8%
  2. 2 Qwen2.5-VL-72B 63.3%
  3. 3 Claude 3.5 Haiku 61.7%
  4. 4 Gemma 3 27B 60.9%
  5. 5 Qwen3-VL-8B-Instruct 54.8%
  6. 6 Gemma 3 12B 52.6%
  7. 7 Qwen2.5-VL-7B 50.9%
  8. 8 Qwen3-VL-30B-A3B-Instruct 50.3%
  9. 9 Qwen3-VL-4B-Instruct 49.5%
  10. 10 Qwen2.5-VL-32B 45.2%
  11. 11 Qwen2.5-VL-3B 42.5%
  12. 12 Idefics3-8B-Llama3 40.9%
  13. 13 llava-v1.6-34b-hf 38.7%
  14. 14 SmolVLM2-256M-Video-Instruct 38.3%
  15. 15 Gemma 3 4B 38.1%
  16. 16 llava-1.5-7b-hf 31.6%
  17. 17 llava-1.5-13b-hf 31.5%
  18. 18 SmolVLM2-2.2B-Instruct 30.4%
  19. 19 SmolVLM2-500M-Video-Instruct 20.5%

Source: Official PISA-Bench leaderboard

#EntryScoreSettingDateLinkSource
1 GPT-4o (snapshot unspecified) 67.8% Official leaderboard
2 Qwen2.5-VL-72B 63.3% Jan 2025 huggingface.co Official leaderboard
3 Claude 3.5 Haiku 61.7% Oct 2024 Official leaderboard
4 Gemma 3 27B 60.9% Mar 2025 huggingface.co Official leaderboard
5 Qwen3-VL-8B-Instruct 54.8% Oct 2025 huggingface.co Official leaderboard
6 Gemma 3 12B 52.6% Mar 2025 huggingface.co Official leaderboard
7 Qwen2.5-VL-7B 50.9% Jan 2025 huggingface.co Official leaderboard
8 Qwen3-VL-30B-A3B-Instruct 50.3% Oct 2025 huggingface.co Official leaderboard
9 Qwen3-VL-4B-Instruct 49.5% Oct 2025 huggingface.co Official leaderboard
10 Qwen2.5-VL-32B 45.2% Mar 2025 huggingface.co Official leaderboard
11 Qwen2.5-VL-3B 42.5% Jan 2025 huggingface.co Official leaderboard
12 Idefics3-8B-Llama3 40.9% Aug 2024 huggingface.co Official leaderboard
13 llava-v1.6-34b-hf 38.7% huggingface.co Official leaderboard
14 SmolVLM2-256M-Video-Instruct 38.3% Feb 2025 huggingface.co Official leaderboard
15 Gemma 3 4B 38.1% Mar 2025 huggingface.co Official leaderboard
16 llava-1.5-7b-hf 31.6% huggingface.co Official leaderboard
17 llava-1.5-13b-hf 31.5% huggingface.co Official leaderboard
18 SmolVLM2-2.2B-Instruct 30.4% Feb 2025 huggingface.co Official leaderboard
19 SmolVLM2-500M-Video-Instruct 20.5% Feb 2025 huggingface.co Official leaderboard
#ModelScoreSettingDateLinkSource
1 Qwen2.5-VL-72B 63.3% Jan 2025 huggingface.co Official leaderboard
2 Claude 3.5 Haiku 61.7% Oct 2024 Official leaderboard
3 Gemma 3 27B 60.9% Mar 2025 huggingface.co Official leaderboard
4 Qwen3-VL-8B-Instruct 54.8% Oct 2025 huggingface.co Official leaderboard
5 Gemma 3 12B 52.6% Mar 2025 huggingface.co Official leaderboard
6 Qwen2.5-VL-7B 50.9% Jan 2025 huggingface.co Official leaderboard
7 Qwen3-VL-30B-A3B-Instruct 50.3% Oct 2025 huggingface.co Official leaderboard
8 Qwen3-VL-4B-Instruct 49.5% Oct 2025 huggingface.co Official leaderboard
9 Qwen2.5-VL-32B 45.2% Mar 2025 huggingface.co Official leaderboard
10 Qwen2.5-VL-3B 42.5% Jan 2025 huggingface.co Official leaderboard
11 Idefics3-8B-Llama3 40.9% Aug 2024 huggingface.co Official leaderboard
12 SmolVLM2-256M-Video-Instruct 38.3% Feb 2025 huggingface.co Official leaderboard
13 Gemma 3 4B 38.1% Mar 2025 huggingface.co Official leaderboard
14 SmolVLM2-2.2B-Instruct 30.4% Feb 2025 huggingface.co Official leaderboard
15 SmolVLM2-500M-Video-Instruct 20.5% Feb 2025 huggingface.co Official leaderboard

Rankings by part

We read 6 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as English, German and French. Qwen2.5-VL-72B leads 3 of them, and 2 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.

English

15 models we track have a score here.

  1. 1 Qwen2.5-VL-72B 69.4%
  2. 2 Claude 3.5 Haiku 62.9%
  3. 3 Gemma 3 27B 60.5%
  4. 4 Qwen3-VL-8B-Instruct 58.9%
  5. 5 Gemma 3 12B 58.1%
  6. 6 Qwen3-VL-30B-A3B-Instruct ↑1 57.3%
  7. 7 Qwen2.5-VL-7B ↓1 52.4%
  8. 8 Qwen2.5-VL-32B ↑1 51.6%
  9. 9 Qwen3-VL-4B-Instruct ↓1 48.4%
  10. 10 Idefics3-8B-Llama3 ↑1 47.6%

Number one over time

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

Qwen
Own leaderboard Not tested since Oct 2025

Source: Official PISA-Bench 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% 2025 Near the ceiling (85%) Qwen2.5-VL-72B Claude 3.5 Haiku Idefics3-8B-Llama3 0% 20% 40% 60% 80% 100% 2025 Near the ceiling (85%) Qwen2.5-VL-72B Idefics3-8B-Llama3

Source: Official PISA-Bench 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 13 months ago, in October 2025, for Qwen3-VL-4B-Instruct. 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 (19 entries).
  • Pass Models scored 15 models we track, plus 4 other entries on its leaderboard.
  • Warning Newest score The newest came 13 months ago, in October 2025, for Qwen3-VL-4B-Instruct.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Gemma 3 27B (#165).
  • Pass Headroom The best is 63.3%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on huggingface.co: 15 rows 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.

How we check