Benchmark Almanac

EduArt

Gemini 3.1 Pro 82.8% leads since July 2026
Not in model cards

Models answer art history questions from Italian secondary schools and US Advanced Placement exams. The authors compare 12 models across seven question formats.

Released
July 2026
Kind
Exam questions
Level
School
Scored
Checked automatically
Answers
Short answer
Tasks
All public
Measures
Capability
Language
Italian, English
In and out
Text, Images → Text
Interaction
Single turn
Items from
Exams and contests
Refresh
One fixed set
Run by
Anyone
In views
Vision
Strongest model tested
GPT-5.5 #21 on the capability index
Last model added
May 2026
Scores refreshed
November 20th, 2026
12
Entries on its own leaderboard, 10 models we track

Gemini 3.1 Pro leads with 82.8% among models released by May 2026 Gemini 3.1 Pro leads with 82.8% among models released by May 2026

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 May 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 Gemini 3.1 Pro Answer only 82.8%
  2. 2 GPT-5.5 Answer only 80%
  3. 3 Gemini 3.5 Flash Answer only 68.9%
  4. 4 Claude Opus 4.6 Answer only 67.3%
  5. 5 Claude Sonnet 4.6 Answer only 64.8%
  6. 6 harvard_qwen.qwen3-vl-235b-a22b Answer only 64.5%
  7. 7 Gemini 3.1 Flash-Lite Answer only 62.8%
  8. 8 Mistral Large 3 Answer only 58.8%
  9. 9 GPT-5.4 Mini Answer only 54.9%
  10. 10 Claude Haiku 4.5 Answer only 53.1%
  11. 11 GPT-5.4 Nano Answer only 40.1%
  12. 12 harvard_us.mistral.pixtral-large-2502-v1:0 Answer only 29.1%

Source: Official EduArt leaderboard

#EntryScoreSettingDateSource
1 Gemini 3.1 Pro 82.8% Answer only Feb 2026 Official leaderboard
2 GPT-5.5 80% Answer only Apr 2026 Official leaderboard
3 Gemini 3.5 Flash 68.9% Answer only May 2026 Official leaderboard
4 Claude Opus 4.6 67.3% Answer only Feb 2026 Official leaderboard
5 Claude Sonnet 4.6 64.8% Answer only Feb 2026 Official leaderboard
6 harvard_qwen.qwen3-vl-235b-a22b 64.5% Answer only Official leaderboard
7 Gemini 3.1 Flash-Lite 62.8% Answer only Mar 2026 Official leaderboard
8 Mistral Large 3 58.8% Answer only Dec 2025 Official leaderboard
9 GPT-5.4 Mini 54.9% Answer only Mar 2026 Official leaderboard
10 Claude Haiku 4.5 53.1% Answer only Oct 2025 Official leaderboard
11 GPT-5.4 Nano 40.1% Answer only Mar 2026 Official leaderboard
12 harvard_us.mistral.pixtral-large-2502-v1:0 29.1% Answer only Official leaderboard
#ModelScoreSettingDateSource
1 Gemini 3.1 Pro 82.8% Answer only Feb 2026 Official leaderboard
2 GPT-5.5 80% Answer only Apr 2026 Official leaderboard
3 Gemini 3.5 Flash 68.9% Answer only May 2026 Official leaderboard
4 Claude Opus 4.6 67.3% Answer only Feb 2026 Official leaderboard
5 Claude Sonnet 4.6 64.8% Answer only Feb 2026 Official leaderboard
6 Gemini 3.1 Flash-Lite 62.8% Answer only Mar 2026 Official leaderboard
7 Mistral Large 3 58.8% Answer only Dec 2025 Official leaderboard
8 GPT-5.4 Mini 54.9% Answer only Mar 2026 Official leaderboard
9 Claude Haiku 4.5 53.1% Answer only Oct 2025 Official leaderboard
10 GPT-5.4 Nano 40.1% Answer only Mar 2026 Official leaderboard

Rankings by part

We read 7 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as MCQ Radio, MCQ Check and True/False. Gemini 3.1 Pro leads 3 of them, and 3 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.

MCQ Radio

10 models we track have a score here.

  1. 1 Gemini 3.5 Flash ↑2 96.5%
  2. 2 Gemini 3.1 Pro ↓1 95.7%
  3. 3 GPT-5.5 ↓1 95.1%
  4. 4 Claude Opus 4.6 94.6%
  5. 5 Claude Sonnet 4.6 94.3%
  6. 6 Gemini 3.1 Flash-Lite 92.4%
  7. 7 Mistral Large 3 84.6%
  8. 8 Claude Haiku 4.5 ↑1 84%
  9. 9 GPT-5.4 Mini ↓1 82.7%
  10. 10 GPT-5.4 Nano 62.7%

Number one over time

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

Google DeepMind
Own leaderboard Not tested since Jul 2026

Source: Official EduArt 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% 2026 Near the ceiling (85%) Gemini 3.1 Pro Claude Opus 4.6 Mistral Large 3 Claude Haiku 4.5 0% 20% 40% 60% 80% 100% 2026 Near the ceiling (85%) Gemini 3.1 Pro Claude Opus 4.6

Source: Official EduArt 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 Knowledge

All 332 in Knowledge
  • MMLU-Pro In use Harder 10-option version of MMLU with reasoning-focused questions across 14 subjects. Intern-S2-397B 89.8%
  • MMLU In use 57-subject multiple-choice exam questions from elementary to professional level. Scored by accuracy. Seed1.8 92.3%
  • SuperGPQA In use 26,529 graduate-level multiple-choice questions across 285 disciplines. Accuracy. Qwen 3.6 Max (Preview) 73.9%
  • WideSearch In use 200 bilingual broad info-seeking tasks: agent collects many atomic facts into a table. Scored by success rate and row/item F1. Tencent Hy4 preview 83.9%
  • BBEH micro average In use Models solve harder replacements for each of the 23 BIG-Bench Hard tasks, 4,520 examples in total. The score is the share of all examples answered correctly. Gemma 4 31B IT 74.4%
  • HLE-Diamond In use Models answer 1,000 cleaned expert questions from Humanity's Last Exam, 500 on reasoning and 500 on knowledge, closed book. Graded on exact answers. Claude Fable 5.2 77.9%

Data audit

9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 6 months ago, in May 2026, for Gemini 3.5 Flash. 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 (12 entries).
  • Pass Models scored 10 models we track, plus 2 other entries on its leaderboard.
  • Warning Newest score The newest came 6 months ago, in May 2026, for Gemini 3.5 Flash.
  • Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5.5 (#21).
  • Pass Headroom The best is 82.8%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on github.com: 10 rows for models we track and 2 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