AECBench (Liang et al.)
Models interpret Chinese building codes, calculate design quantities, and write project documents. Answer keys score objective questions. A model judge grades report generation against expert rubrics.
The authors publish nine model scores for each of 23 tasks. The paper counts 4,800 questions. Seven public task files double their reported counts.
- Released
- September 2025
- Kind
- Expert-written questions
- Level
- University
- Scored
- Checked automatically
- Answers
- Multiple choice
- Tasks
- All public
- Measures
- Capability
- Subject
- Civil engineering
- Language
- Chinese
- Interaction
- Single turn
- Items from
- Written by experts
- Refresh
- One fixed set
- Run by
- Anyone
- Strongest model tested
- o3-mini #124 on the capability index
- Last model added
- March 2025
- Scores refreshed
- November 20th, 2026
DeepSeek-R1 (snapshot unspecified) tops the leaderboard with 70.5 Hunyuan-TurboS leads with 61.6 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. The best model we track is Hunyuan-TurboS, at 61.6. 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:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 16 more
- 1 DeepSeek-R1 (snapshot unspecified) 70.5
- 2 DeepSeek-V3 (snapshot unspecified) 65.8
- 3
Hunyuan-TurboS 61.6
- 4 o3-mini 59.9
- 5 GLM-4-Plus 59.0
- 6
QwQ-32B 58.9
- 7
Qwen-Turbo 58.6
- 8 GPT-4o (snapshot unspecified) 57.5
- 9 Moonshot-v1-128k 52.4
Source: Official AECBench (Liang et al.) leaderboard
- 1
Hunyuan-TurboS 61.6
- 2 o3-mini 59.9
- 3
QwQ-32B 58.9
- 4
Qwen-Turbo 58.6
Source: Official AECBench (Liang et al.) leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | DeepSeek-R1 (snapshot unspecified) | 70.5 | Official leaderboard | ||
| 2 | DeepSeek-V3 (snapshot unspecified) | 65.8 | Official leaderboard | ||
| 3 | | 61.6 | Feb 2025 | Official leaderboard | |
| 4 | o3-mini | 59.9 | Jan 2025 | Official leaderboard | |
| 5 | GLM-4-Plus | 59.0 | Official leaderboard | ||
| 6 | | 58.9 | Mar 2025 | Official leaderboard | |
| 7 | | 58.6 | Nov 2024 | Official leaderboard | |
| 8 | GPT-4o (snapshot unspecified) | 57.5 | Official leaderboard | ||
| 9 | Moonshot-v1-128k | 52.4 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 61.6 | Feb 2025 | Official leaderboard | |
| 2 | o3-mini | 59.9 | Jan 2025 | Official leaderboard | |
| 3 | | 58.9 | Mar 2025 | Official leaderboard | |
| 4 | | 58.6 | Nov 2024 | Official leaderboard |
Rankings by part
We read 23 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as Code Memorization & Retrieval, Terminology Memorization and Abbreviation Memorization. Hunyuan-TurboS leads 11 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.
Code Memorization & Retrieval
4 models we track have a score here.
- 1
Hunyuan-TurboS 82.7
- 2
Qwen-Turbo ↑2 70.5
- 3
QwQ-32B 69.5
- 4 o3-mini ↓2 63.6
Terminology Memorization
4 models we track have a score here.
- 1
Qwen-Turbo ↑3 98.5
- 2
Hunyuan-TurboS ↓1 98.5
- 3
QwQ-32B 97.0
- 4 o3-mini ↓2 95.0
Abbreviation Memorization
4 models we track have a score here.
- 1 o3-mini ↑1 86.1
- 2
QwQ-32B ↑1 77.2
- 3
Hunyuan-TurboS ↓2 75.0
- 4
Qwen-Turbo 67.8
Code Provision Interpretation
4 models we track have a score here.
- 1
Hunyuan-TurboS 77.0
- 2
Qwen-Turbo ↑2 73.4
- 3
QwQ-32B 67.7
- 4 o3-mini ↓2 54.1
Design Literacy Q&A
4 models we track have a score here.
- 1
Hunyuan-TurboS 90.3
- 2
QwQ-32B ↑1 88.5
- 3
Qwen-Turbo ↑1 87.9
- 4 o3-mini ↓2 70.9
Code Provision Interpretation (Tabular Data)
4 models we track have a score here.
- 1
Qwen-Turbo ↑3 43.9
- 2
Hunyuan-TurboS ↓1 43.9
- 3
QwQ-32B 40.7
- 4 o3-mini ↓2 39.7
Design Decision Formulation
4 models we track have a score here.
- 1
Qwen-Turbo ↑3 74.6
- 2
Hunyuan-TurboS ↓1 74.6
- 3
QwQ-32B 71.4
- 4 o3-mini ↓2 60.5
Building Type Inference
4 models we track have a score here.
- 1
Hunyuan-TurboS 88.0
- 2
QwQ-32B ↑1 83.0
- 3 o3-mini ↓1 82.5
- 4
Qwen-Turbo 76.0
Design Decision Formulation (Tabular Data)
4 models we track have a score here.
- 1
Hunyuan-TurboS 44.5
- 2
Qwen-Turbo ↑2 43.3
- 3 o3-mini ↓1 42.9
- 4
QwQ-32B ↓1 42.1
Calculation in Architectural Design
4 models we track have a score here.
- 1 o3-mini ↑1 51.3
- 2
QwQ-32B ↑1 43.9
- 3
Qwen-Turbo ↑1 41.3
- 4
Hunyuan-TurboS ↓3 33.9
Calculation in MEP Design
4 models we track have a score here.
- 1 o3-mini ↑1 92.3
- 2
QwQ-32B ↑1 88.5
- 3
Qwen-Turbo ↑1 69.2
- 4
Hunyuan-TurboS ↓3 41.5
Calculation in Construction
4 models we track have a score here.
- 1
QwQ-32B ↑2 83.6
- 2 o3-mini 75.4
- 3
Qwen-Turbo ↑1 65.6
- 4
Hunyuan-TurboS ↓3 48.1
Calculation in Engineering Economics
4 models we track have a score here.
- 1
Hunyuan-TurboS 67.2
- 2
Qwen-Turbo ↑2 50.9
- 3 o3-mini ↓1 44.1
- 4
QwQ-32B ↓1 25.4
Document Classification
4 models we track have a score here.
- 1 o3-mini ↑1 87.2
- 2
Hunyuan-TurboS ↓1 85.0
- 3
Qwen-Turbo ↑1 77.2
- 4
QwQ-32B ↓1 77.0
Document Proofreading
4 models we track have a score here.
- 1
Qwen-Turbo ↑3 93.8
- 2 o3-mini 89.5
- 3
QwQ-32B 86.1
- 4
Hunyuan-TurboS ↓3 81.2
Compliance Checking
4 models we track have a score here.
- 1 o3-mini ↑1 48.7
- 2
Hunyuan-TurboS ↓1 45.2
- 3
QwQ-32B 45.2
- 4
Qwen-Turbo 38.7
Brand Compliance Verification
4 models we track have a score here.
- 1
Hunyuan-TurboS 37.0
- 2 o3-mini 29.9
- 3
Qwen-Turbo ↑1 28.4
- 4
QwQ-32B ↓1 18.8
Information Extraction
4 models we track have a score here.
- 1
Hunyuan-TurboS 73.1
- 2 o3-mini 69.1
- 3
Qwen-Turbo ↑1 67.3
- 4
QwQ-32B ↓1 44.1
Evaluation of Documents in Architectural Design
4 models we track have a score here.
- 1
Hunyuan-TurboS 54.9
- 2 o3-mini 50.5
- 3
Qwen-Turbo ↑1 47.4
- 4
QwQ-32B ↓1 38.3
Evaluation of Documents in Structural Design
4 models we track have a score here.
- 1
QwQ-32B ↑2 51.1
- 2
Hunyuan-TurboS ↓1 49.6
- 3 o3-mini ↓1 46.6
- 4
Qwen-Turbo 41.7
Conceptual Design Proposal Generation
4 models we track have a score here.
- 1
Hunyuan-TurboS 43.9
- 2
QwQ-32B ↑1 32.7
- 3 o3-mini ↓1 29.1
- 4
Qwen-Turbo 24.1
Specialized Report Generation
4 models we track have a score here.
- 1
Hunyuan-TurboS 40.0
- 2
Qwen-Turbo ↑2 33.3
- 3
QwQ-32B 30.0
- 4 o3-mini ↓2 15.6
Tender Review Report Generation
4 models we track have a score here.
- 1 o3-mini ↑1 53.3
- 2
QwQ-32B ↑1 52.8
- 3
Hunyuan-TurboS ↓2 42.8
- 4
Qwen-Turbo 33.6
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official AECBench (Liang et al.) leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official AECBench (Liang et al.) 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
- 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
8 of 9 checks pass, 1 warning. Look first at the Newest score check. The newest came 20 months ago, in March 2025, for QwQ-32B. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 2 pass 2 notes
- Pass Independent runs Scores from the official leaderboard (9 entries).
- Pass Models scored 4 models we track, plus 5 other entries on its leaderboard.
- Warning Newest score The newest came 20 months ago, in March 2025, for QwQ-32B.
- Note Frontier models None of the top 10 has a score. The strongest tested is o3-mini (#124).
- Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on github.com: 4 rows 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.