Benchmark Almanac

AECV-Bench (2026 release)

Gemini 3.5 Flash 71.4% leads since May 2026
Not in model cards

Models count four types of floor plan elements and answer questions about engineering drawings. Counting uses exact matches. A model grades the drawing answers.

AECFoundry ranks 12 models in each track. This release covers 120 floor plans and 192 questions across 21 drawings.

Made by
AECFoundry
Released
December 2025
Kind
Documents, charts and text in images
Level
University
Scored
Graded by a model
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
In views
Vision
Strongest model tested
Claude Fable 5 #12 on the capability index
Last model added
June 2026
Scores refreshed
November 20th, 2026
12
Entries on its own leaderboard, 11 models we track

Gemini 3.5 Flash leads with 71.4% among models released by June 2026 Gemini 3.5 Flash leads with 71.4% among models released by June 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 June 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 3.5 Flash 71.4%
  2. 2 Claude Fable 5 70.8%
  3. 3 GPT-5.5 67.7%
  4. 4 Qwen3.7-Plus 63.9%
  5. 5 Kimi K2.6 61.8%
  6. 6 MiniMax-M3 61.4%
  7. 7 StepFun Step 3.7 55.4%
  8. 8 GLM-4.6V 55.2%
  9. 9 Grok 4.3 53.9%
  10. 10 Mistral Large 3 45.5%
  11. 11 Nova 2 Lite 45.4%
  12. 12 Cohere Command A 42.4%

Source: Official AECV-Bench (2026 release) leaderboard

#EntryScoreSettingDateSource
1 Gemini 3.5 Flash 71.4% May 2026 Official leaderboard
2 Claude Fable 5 70.8% Jun 2026 Official leaderboard
3 GPT-5.5 67.7% Apr 2026 Official leaderboard
4 Qwen3.7-Plus 63.9% Jun 2026 Official leaderboard
5 Kimi K2.6 61.8% Apr 2026 Official leaderboard
6 MiniMax-M3 61.4% Jun 2026 Official leaderboard
7 StepFun Step 3.7 55.4% Official leaderboard
8 GLM-4.6V 55.2% Dec 2025 Official leaderboard
9 Grok 4.3 53.9% Apr 2026 Official leaderboard
10 Mistral Large 3 45.5% Dec 2025 Official leaderboard
11 Nova 2 Lite 45.4% Dec 2025 Official leaderboard
12 Cohere Command A 42.4% Mar 2025 Official leaderboard
#ModelScoreSettingDateSource
1 Gemini 3.5 Flash 71.4% May 2026 Official leaderboard
2 Claude Fable 5 70.8% Jun 2026 Official leaderboard
3 GPT-5.5 67.7% Apr 2026 Official leaderboard
4 Qwen3.7-Plus 63.9% Jun 2026 Official leaderboard
5 Kimi K2.6 61.8% Apr 2026 Official leaderboard
6 MiniMax-M3 61.4% Jun 2026 Official leaderboard
7 GLM-4.6V 55.2% Dec 2025 Official leaderboard
8 Grok 4.3 53.9% Apr 2026 Official leaderboard
9 Mistral Large 3 45.5% Dec 2025 Official leaderboard
10 Nova 2 Lite 45.4% Dec 2025 Official leaderboard
11 Cohere Command A 42.4% Mar 2025 Official leaderboard

Rankings by part

We read 10 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as Object Counting, Document QA and Door. Gemini 3.5 Flash leads 7 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.

Object Counting

11 models we track have a score here.

  1. 1 Gemini 3.5 Flash 54.3%
  2. 2 Claude Fable 5 51.5%
  3. 3 GPT-5.5 48.3%
  4. 4 MiniMax-M3 ↑2 47.3%
  5. 5 Qwen3.7-Plus ↓1 46%
  6. 6 Kimi K2.6 ↓1 44.4%
  7. 7 GLM-4.6V 39%
  8. 8 Grok 4.3 35.8%
  9. 9 Nova 2 Lite ↑1 33.5%
  10. 10 Mistral Large 3 ↓1 31.7%

6 models have held number one on the official leaderboard since its release in December 2025

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

Cohere Mistral AI Z.ai Moonshot AI OpenAI Google DeepMind
Own leaderboard Not tested since Jun 2026

Source: Official AECV-Bench (2026 release) 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 2026 Near the ceiling (85%) Gemini 3.5 Flash Kimi K2.6 GLM-4.6V Mistral Large 3 Cohere Command A 0% 20% 40% 60% 80% 100% 2025 2026 Near the ceiling (85%) Gemini 3.5 Flash Kimi K2.6

Source: Official AECV-Bench (2026 release) 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 10 checks pass. Every benchmark

Catalog record 2 pass
  • Pass Task access All tasks are public.
  • Pass Confidence Checked against a primary source.
Scores 4 pass 1 note
  • Pass Independent runs Scores from the official leaderboard (12 entries).
  • Pass Models scored 11 models we track, plus 1 other entry on its leaderboard.
  • Pass Newest score The newest is from June 2026, for Claude Fable 5.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Claude Fable 5 (#12).
  • Pass Headroom The best is 71.4%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on aecv-bench.com: 11 rows for models we track and 1 other entry 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