Benchmark Almanac

Construction AI Leaderboard

Claude Opus 5.5 73.7 leads since November 2026
Not in model cards

Models answer construction questions, inspect drawings, edit building models, and generate validation rules. ONESTRUCTION averages the representative scores from five evaluations.

ONESTRUCTION runs each model once per benchmark. Its BIM editing grader and turn limit differ from the original protocol.

Made by
ONESTRUCTION
Released
November 2026
Kind
Composite indices
Level
University
Scored
Checked automatically
Answers
Agent task
Tasks
All public
Measures
Capability
Language
English
In and out
Text, Documents, 3D → Actions, Text, Code, 3D
Interaction
Agentic
Environment
APIs and tools
Items from
Another benchmark
Refresh
One fixed set
Run by
A host
Strongest model tested
GPT-6 Astra #4 on the capability index
Last model added
October 2026
Scores refreshed
November 20th, 2026
20
Models scored on its own leaderboard

Claude Opus 5.5 leads with 73.7

Best score per model from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best.

  1. 1 Claude Opus 5.5 medium 73.7
  2. 2 Gemini 4 Argon medium 73.6
  3. 3 Claude Opus 5 medium 72.7
  4. 4 Claude Fable 5 medium 70.2
  5. 5 GPT-6.1 Sol medium 70.2
  6. 6 GPT-6 Astra medium 70.1
  7. 7 Grok 4.7 medium 68.7
  8. 8 Muse Spark 1.3 medium 67.2
  9. 9 Claude Sonnet 5.5 medium 66.9
  10. 10 DeepSeek V4.1 Flash 1017 66.7
  11. 11 Gemini 3.7 Flash medium 66.5
  12. 12 Kimi K3 65.7
  13. 13 GLM-5.3-Flash 64.2
  14. 14 Claude Sonnet 5 medium 61.2
  15. 15 DeepSeek V4.1 Flash 59.4
  16. 16 Gemini 3.1 Pro medium 59.2
  17. 17 Qwen 3.8 27B 58.1
  18. 18 Muse Glimmer 49.6
  19. 19 Gemma 4 31B IT 40.8
  20. 20 Nemotron 3 Nano Omni 30B-A3B 26.5

Source: Official Construction AI Leaderboard leaderboard

#ModelScoreSettingDateLinkSource
1 Claude Opus 5.5 73.7 medium, pi Sep 2026 ai.onestruction.com Official leaderboard
2 Gemini 4 Argon 73.6 medium, pi Sep 2026 ai.onestruction.com Official leaderboard
3 Claude Opus 5 72.7 medium, pi Jul 2026 ai.onestruction.com Official leaderboard
4 Claude Fable 5 70.2 medium, pi Jun 2026 ai.onestruction.com Official leaderboard
5 GPT-6.1 Sol 70.2 medium, pi Sep 2026 ai.onestruction.com Official leaderboard
6 GPT-6 Astra 70.1 medium, pi Sep 2026 ai.onestruction.com Official leaderboard
7 Grok 4.7 68.7 medium, pi Sep 2026 ai.onestruction.com Official leaderboard
8 Muse Spark 1.3 67.2 medium, pi Sep 2026 ai.onestruction.com Official leaderboard
9 Claude Sonnet 5.5 66.9 medium, pi Sep 2026 ai.onestruction.com Official leaderboard
10 DeepSeek V4.1 Flash 1017 66.7 pi Oct 2026 ai.onestruction.com Official leaderboard
11 Gemini 3.7 Flash 66.5 medium, pi Aug 2026 ai.onestruction.com Official leaderboard
12 Kimi K3 65.7 pi Jul 2026 ai.onestruction.com Official leaderboard
13 GLM-5.3-Flash 64.2 pi Aug 2026 ai.onestruction.com Official leaderboard
14 Claude Sonnet 5 61.2 medium, pi Jun 2026 ai.onestruction.com Official leaderboard
15 DeepSeek V4.1 Flash 59.4 pi Sep 2026 ai.onestruction.com Official leaderboard
16 Gemini 3.1 Pro 59.2 medium, pi Feb 2026 ai.onestruction.com Official leaderboard
17 Qwen 3.8 27B 58.1 pi Aug 2026 ai.onestruction.com Official leaderboard
18 Muse Glimmer 49.6 pi Aug 2026 ai.onestruction.com Official leaderboard
19 Gemma 4 31B IT 40.8 pi Apr 2026 ai.onestruction.com Official leaderboard
20 Nemotron 3 Nano Omni 30B-A3B 26.5 pi Apr 2026 ai.onestruction.com Official leaderboard

Rankings by part

We read 5 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as IDS generation, Drawings & documents and BIM understanding. Claude Opus 5.5 leads 2 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.

IDS generation

20 models we track have a score here.

  1. 1 Claude Opus 5.5 63.0
  2. 2 Gemini 4 Argon 61.9
  3. 3 GPT-6 Astra ↑3 61.7
  4. 4 Claude Opus 5 ↓1 60.9
  5. 5 GPT-6.1 Sol 60.7
  6. 6 Claude Sonnet 5.5 ↑3 60.0
  7. 7 Claude Fable 5 ↓3 58.6
  8. 8 Grok 4.7 ↓1 58.2
  9. 9 Gemini 3.7 Flash ↑2 52.5
  10. 10 Gemini 3.1 Pro ↑6 50.1

Number one over time

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

Anthropic
Own leaderboard 1 leader

Source: Official Construction AI Leaderboard 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
20 40 60 80 2026 20 40 60 80 2026

Source: Official Construction AI Leaderboard 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 Other

All 69 in Other
  • Artificial Analysis Intelligence Index v5 In use Version 5 of Artificial Analysis's index: nine evaluations in four equal categories, most with private test sets, run three times each in its own agent harness. Claude Fable 5.2 64.6
  • Personal Superintelligence Index (Meta) In use Meta's composite of seven personal and knowledge-work evals: Agents' Last Exam, AutomationBench-AA, FocusBench, JobBench, GDP.pdf, VBT and DRACO.
  • Vals Science Index In use Agents are scored on the equal average of BioMysteryBench, MysteryMechanism and Terminal-Bench-Science. GPT-6.2 Astra 68.4%
  • Decision Index Public v0.3 In use Decision models return probabilities over labeled options in public tasks covering knowledge, language, retrieval, tools, and arts. A chance-corrected weighted aggregate combines the task scores. Granite Switch 4.1 30B 51.3
  • SAGE v1.1 (Vals) In use Models grade handwritten undergraduate math and computer science work against rubrics. Scored by class-balanced accuracy of each rubric check against expert grading. Claude Opus 5.5 91.3%
  • MTEB Fading Massive Text Embedding Benchmark: retrieval, clustering, classification across languages and modalities. Qwen3-Embedding-8B 72.2%

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 (20 entries).
  • Pass Models scored 20 models we track.
  • Pass Newest score The newest is from October 2026, for DeepSeek V4.1 Flash 1017.
  • Pass Frontier models 5 of the top 10 have a score, led by GPT-6 Astra (#4).
  • Note Headroom Scored as an index, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on ai.onestruction.com: 20 rows for models we track 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