Benchmark Almanac

AECBench (Liang et al.)

DeepSeek-R1 (snapshot unspecified) 70.5 tops its leaderboard Hunyuan-TurboS 61.6 leads the models we track
Not in model cards

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
Language
Chinese
In and out
Text → Text
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
9
Entries on its own leaderboard, 4 models we track

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:

  1. 1 DeepSeek-R1 (snapshot unspecified) 70.5
  2. 2 DeepSeek-V3 (snapshot unspecified) 65.8
  3. 3 Hunyuan-TurboS 61.6
  4. 4 o3-mini 59.9
  5. 5 GLM-4-Plus 59.0
  6. 6 QwQ-32B 58.9
  7. 7 Qwen-Turbo 58.6
  8. 8 GPT-4o (snapshot unspecified) 57.5
  9. 9 Moonshot-v1-128k 52.4

Source: Official AECBench (Liang et al.) leaderboard

Source: Official AECBench (Liang et al.) leaderboard

#EntryScoreSettingDateSource
1 DeepSeek-R1 (snapshot unspecified) 70.5 Official leaderboard
2 DeepSeek-V3 (snapshot unspecified) 65.8 Official leaderboard
3 Hunyuan-TurboS 61.6 Feb 2025 Official leaderboard
4 o3-mini 59.9 Jan 2025 Official leaderboard
5 GLM-4-Plus 59.0 Official leaderboard
6 QwQ-32B 58.9 Mar 2025 Official leaderboard
7 Qwen-Turbo 58.6 Nov 2024 Official leaderboard
8 GPT-4o (snapshot unspecified) 57.5 Official leaderboard
9 Moonshot-v1-128k 52.4 Official leaderboard
#ModelScoreSettingDateSource
1 Hunyuan-TurboS 61.6 Feb 2025 Official leaderboard
2 o3-mini 59.9 Jan 2025 Official leaderboard
3 QwQ-32B 58.9 Mar 2025 Official leaderboard
4 Qwen-Turbo 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. 1 Hunyuan-TurboS 82.7
  2. 2 Qwen-Turbo ↑2 70.5
  3. 3 QwQ-32B 69.5
  4. 4 o3-mini ↓2 63.6

Number one over time

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

Tencent
Own leaderboard Not tested since Sep 2025

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.

Plot with other benchmarks
On the benchmark’s own leaderboard
58.0 59.0 60.0 61.0 62.0 2025 58.0 59.0 60.0 61.0 62.0 2025

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

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

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.

How we check