Construction's Last Exam
Version geometry-only
Models trace walls and areas and locate symbols on construction drawings. The score averages the three task families across a suite of 33 tasks.
Togal ranks 31 models and lists its human-assisted software separately. Eleven model scores differ between the ranking table and the progress chart.
- Made by
- Togal.AI
- Released
- October 2026
- Kind
- Tasks from real jobs
- Level
- University
- Scored
- Checked automatically
- Answers
- Agent task
- Tasks
- All public
- Measures
- Capability
- Subject
- Civil engineering
- Language
- English
- Interaction
- Single turn
- Refresh
- One fixed set
- Run by
- Anyone
- Strongest model tested
- GPT-6.2 Astra #1 on the capability index
- Last model added
- November 2026
- Scores refreshed
- November 20th, 2026
Togal.ai tops the leaderboard with 97.7% GPT-6.2 Astra leads with 58.1%
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. Every run is in the next chart. 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 GPT-6.2 Astra, at 58.1%. This view shows only models we track, each at its best.
- 97.7 Togal.ai
- 58.1 GPT-6.2 Astra
- 44.2 GPT-6.2 Sol
- 36.4 GPT-6.1 Terra
- 34.0 GPT-6 Sol
- 28.3
Gemini 4 Neon
- 25.7
Gemini 4 Argon
- 21.6 GPT-6 Luna
- 19.2 GPT-6 Astra
- 19.1
Gemini 3.8 Flash
- 16.7
Gemini 4 Hydrogen
- 16.6
Gemini 3.7 Flash
- 16.2 Claude Opus 5.5
- 12.7 Claude Opus 5
- 12.5 Claude Fable 5.1
- 12.3 DeepSeek Flash 4.1
- 12.0
Gemini 3.6 Flash
- 11.1 GPT-5.6 Sol
- 10.1 GPT-5.6 Luna
- 6.00
Gemini 3.1 Flash-Lite
- 5.50 GLM-5.3-Flash
- 4.90
Gemini 3.5 Flash-Lite
- 4.00
MiniMax-M3.1-Flash
- 3.80
Gemini 2.5 Flash
- 3.40 Claude Opus 4.6
- 3.30 Claude Haiku 4.5
- 2.20 Claude Sonnet 4.5
- 2.20
DeepSeek V4.1 Flash 1017
- 2.10 Claude Sonnet 5
- 1.80
Qwen2.5-VL-72B
- 1.30 GLM-4.6V
- 1.30
Qwen 3.8 Max
- 1 Togal.ai Human-assisted software 97.7%
- 2 GPT-6.2 Astra 58.1%
- 3 GPT-6.2 Sol 44.2%
- 4 GPT-6.1 Terra 36.4%
- 5 GPT-6 Sol 34%
- 6
Gemini 4 Neon 28.3%
- 7
Gemini 4 Argon 25.7%
- 8 GPT-6 Luna 21.6%
- 9 GPT-6 Astra 19.2%
- 10
Gemini 3.8 Flash 19.1%
- 11
Gemini 4 Hydrogen 16.7%
- 12
Gemini 3.7 Flash 16.6%
- 13 Claude Opus 5.5 16.2%
- 14 Claude Opus 5 12.7%
- 15 Claude Fable 5.1 12.5%
- 16 DeepSeek Flash 4.1 12.3%
- 17
Gemini 3.6 Flash 12%
- 18 GPT-5.6 Sol 11.1%
- 19 GPT-5.6 Luna 10.1%
- 20
Gemini 3.1 Flash-Lite 6.00%
- 21 GLM-5.3-Flash 5.50%
- 22
Gemini 3.5 Flash-Lite 4.90%
- 23
MiniMax-M3.1-Flash 4.00%
- 24
Gemini 2.5 Flash Sep 2025 3.80%
- 25 Claude Opus 4.6 3.40%
- 26 Claude Haiku 4.5 3.30%
- 27 Claude Sonnet 4.5 2.20%
- 28
DeepSeek V4.1 Flash 1017 2.20%
- 29 Claude Sonnet 5 2.10%
- 30
Qwen2.5-VL-72B 1.80%
- 31 GLM-4.6V 1.30%
- 32
Qwen 3.8 Max 1.30%
Source: Official Construction's Last Exam leaderboard
- 58.1 GPT-6.2 Astra
- 44.2 GPT-6.2 Sol
- 36.4 GPT-6.1 Terra
- 34.0 GPT-6 Sol
- 28.3
Gemini 4 Neon
- 25.7
Gemini 4 Argon
- 21.6 GPT-6 Luna
- 19.2 GPT-6 Astra
- 19.1
Gemini 3.8 Flash
- 16.7
Gemini 4 Hydrogen
- 16.6
Gemini 3.7 Flash
- 16.2 Claude Opus 5.5
- 12.7 Claude Opus 5
- 12.5 Claude Fable 5.1
- 12.0
Gemini 3.6 Flash
- 11.1 GPT-5.6 Sol
- 10.1 GPT-5.6 Luna
- 6.00
Gemini 3.1 Flash-Lite
- 5.50 GLM-5.3-Flash
- 4.90
Gemini 3.5 Flash-Lite
- 4.00
MiniMax-M3.1-Flash
- 3.80
Gemini 2.5 Flash
- 3.40 Claude Opus 4.6
- 3.30 Claude Haiku 4.5
- 2.20 Claude Sonnet 4.5
- 2.20
DeepSeek V4.1 Flash 1017
- 2.10 Claude Sonnet 5
- 1.80
Qwen2.5-VL-72B
- 1.30 GLM-4.6V
- 1.30
Qwen 3.8 Max
- 1 GPT-6.2 Astra 58.1%
- 2 GPT-6.2 Sol 44.2%
- 3 GPT-6.1 Terra 36.4%
- 4 GPT-6 Sol 34%
- 5
Gemini 4 Neon 28.3%
- 6
Gemini 4 Argon 25.7%
- 7 GPT-6 Luna 21.6%
- 8 GPT-6 Astra 19.2%
- 9
Gemini 3.8 Flash 19.1%
- 10
Gemini 4 Hydrogen 16.7%
- 11
Gemini 3.7 Flash 16.6%
- 12 Claude Opus 5.5 16.2%
- 13 Claude Opus 5 12.7%
- 14 Claude Fable 5.1 12.5%
- 15
Gemini 3.6 Flash 12%
- 16 GPT-5.6 Sol 11.1%
- 17 GPT-5.6 Luna 10.1%
- 18
Gemini 3.1 Flash-Lite 6.00%
- 19 GLM-5.3-Flash 5.50%
- 20
Gemini 3.5 Flash-Lite 4.90%
- 21
MiniMax-M3.1-Flash 4.00%
- 22
Gemini 2.5 Flash Sep 2025 3.80%
- 23 Claude Opus 4.6 3.40%
- 24 Claude Haiku 4.5 3.30%
- 25 Claude Sonnet 4.5 2.20%
- 26
DeepSeek V4.1 Flash 1017 2.20%
- 27 Claude Sonnet 5 2.10%
- 28
Qwen2.5-VL-72B 1.80%
- 29 GLM-4.6V 1.30%
- 30
Qwen 3.8 Max 1.30%
Source: Official Construction's Last Exam leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Togal.ai | 97.7% | Human-assisted software | Official leaderboard | |
| 2 | GPT-6.2 Astra | 58.1% | Nov 2026 | Official leaderboard | |
| 3 | GPT-6.2 Sol | 44.2% | Nov 2026 | Official leaderboard | |
| 4 | GPT-6.1 Terra | 36.4% | Oct 2026 | Official leaderboard | |
| 5 | GPT-6 Sol | 34% | Sep 2026 | Official leaderboard | |
| 6 | | 28.3% | Nov 2026 | Official leaderboard | |
| 7 | | 25.7% | Sep 2026 | Official leaderboard | |
| 8 | GPT-6 Luna | 21.6% | Sep 2026 | Official leaderboard | |
| 9 | GPT-6 Astra | 19.2% | Sep 2026 | Official leaderboard | |
| 10 | | 19.1% | Sep 2026 | Official leaderboard | |
| 11 | | 16.7% | Nov 2026 | Official leaderboard | |
| 12 | | 16.6% | Aug 2026 | Official leaderboard | |
| 13 | Claude Opus 5.5 | 16.2% | Sep 2026 | Official leaderboard | |
| 14 | Claude Opus 5 | 12.7% | Jul 2026 | Official leaderboard | |
| 15 | Claude Fable 5.1 | 12.5% | Sep 2026 | Official leaderboard | |
| 16 | DeepSeek Flash 4.1 | 12.3% | Official leaderboard | ||
| 17 | | 12% | Jul 2026 | Official leaderboard | |
| 18 | GPT-5.6 Sol | 11.1% | Jul 2026 | Official leaderboard | |
| 19 | GPT-5.6 Luna | 10.1% | Jul 2026 | Official leaderboard | |
| 20 | | 6.00% | Mar 2026 | Official leaderboard | |
| 21 | GLM-5.3-Flash | 5.50% | Aug 2026 | Official leaderboard | |
| 22 | | 4.90% | Jul 2026 | Official leaderboard | |
| 23 | | 4.00% | Oct 2026 | Official leaderboard | |
| 24 | | 3.80% | Sep 2025 | Official leaderboard | |
| 25 | Claude Opus 4.6 | 3.40% | Feb 2026 | Official leaderboard | |
| 26 | Claude Haiku 4.5 | 3.30% | Oct 2025 | Official leaderboard | |
| 27 | Claude Sonnet 4.5 | 2.20% | Sep 2025 | Official leaderboard | |
| 28 | | 2.20% | Oct 2026 | Official leaderboard | |
| 29 | Claude Sonnet 5 | 2.10% | Jun 2026 | Official leaderboard | |
| 30 | | 1.80% | Jan 2025 | Official leaderboard | |
| 31 | GLM-4.6V | 1.30% | Dec 2025 | Official leaderboard | |
| 32 | | 1.30% | Aug 2026 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-6.2 Astra | 58.1% | Nov 2026 | Official leaderboard | |
| 2 | GPT-6.2 Sol | 44.2% | Nov 2026 | Official leaderboard | |
| 3 | GPT-6.1 Terra | 36.4% | Oct 2026 | Official leaderboard | |
| 4 | GPT-6 Sol | 34% | Sep 2026 | Official leaderboard | |
| 5 | | 28.3% | Nov 2026 | Official leaderboard | |
| 6 | | 25.7% | Sep 2026 | Official leaderboard | |
| 7 | GPT-6 Luna | 21.6% | Sep 2026 | Official leaderboard | |
| 8 | GPT-6 Astra | 19.2% | Sep 2026 | Official leaderboard | |
| 9 | | 19.1% | Sep 2026 | Official leaderboard | |
| 10 | | 16.7% | Nov 2026 | Official leaderboard | |
| 11 | | 16.6% | Aug 2026 | Official leaderboard | |
| 12 | Claude Opus 5.5 | 16.2% | Sep 2026 | Official leaderboard | |
| 13 | Claude Opus 5 | 12.7% | Jul 2026 | Official leaderboard | |
| 14 | Claude Fable 5.1 | 12.5% | Sep 2026 | Official leaderboard | |
| 15 | | 12% | Jul 2026 | Official leaderboard | |
| 16 | GPT-5.6 Sol | 11.1% | Jul 2026 | Official leaderboard | |
| 17 | GPT-5.6 Luna | 10.1% | Jul 2026 | Official leaderboard | |
| 18 | | 6.00% | Mar 2026 | Official leaderboard | |
| 19 | GLM-5.3-Flash | 5.50% | Aug 2026 | Official leaderboard | |
| 20 | | 4.90% | Jul 2026 | Official leaderboard | |
| 21 | | 4.00% | Oct 2026 | Official leaderboard | |
| 22 | | 3.80% | Sep 2025 | Official leaderboard | |
| 23 | Claude Opus 4.6 | 3.40% | Feb 2026 | Official leaderboard | |
| 24 | Claude Haiku 4.5 | 3.30% | Oct 2025 | Official leaderboard | |
| 25 | Claude Sonnet 4.5 | 2.20% | Sep 2025 | Official leaderboard | |
| 26 | | 2.20% | Oct 2026 | Official leaderboard | |
| 27 | Claude Sonnet 5 | 2.10% | Jun 2026 | Official leaderboard | |
| 28 | | 1.80% | Jan 2025 | Official leaderboard | |
| 29 | GLM-4.6V | 1.30% | Dec 2025 | Official leaderboard | |
| 30 | | 1.30% | Aug 2026 | Official leaderboard |
Rankings by part
We read 3 parts from the benchmark’s own leaderboard: Linear, Areas and Counting. Each ranks models on one part of the test. GPT-6.2 Astra leads every one of them. Arrows show how far a model moved from its overall rank among the models we track.
Linear
29 models we track have a score here.
- 1 GPT-6.2 Astra 51.2%
- 2 GPT-6 Sol ↑2 30.9%
- 3
Gemini 4 Neon ↑2 28.5%
- 4 GPT-6.2 Sol ↓2 27.2%
- 5
Gemini 4 Argon ↑1 23.5%
- 6 GPT-6.1 Terra ↓3 22.1%
- 7 GPT-6 Astra ↑1 21.8%
- 8
Gemini 3.8 Flash ↑1 18.7%
- 9 GPT-6 Luna ↓2 15.3%
- 10 Claude Opus 5 ↑3 14.8%
Areas
29 models we track have a score here.
- 1 GPT-6.2 Astra 47.8%
- 2 GPT-6 Sol ↑2 29.8%
- 3 GPT-6.2 Sol ↓1 28.6%
- 4 GPT-6.1 Terra ↓1 27%
- 5
Gemini 4 Neon 23.9%
- 6 GPT-6 Luna ↑1 21.3%
- 7 Claude Opus 5.5 ↑5 20.6%
- 8
Gemini 4 Argon ↓2 20.5%
- 9
Gemini 3.7 Flash ↑2 18.5%
- 10
Gemini 4 Hydrogen 18.2%
Counting
29 models we track have a score here.
- 1 GPT-6.2 Astra 71.7%
- 2 GPT-6.2 Sol 69.9%
- 3 GPT-6.1 Terra 54.9%
- 4 GPT-6 Sol 40.5%
- 5
Gemini 4 Neon 31.9%
- 6
Gemini 4 Argon 31.8%
- 7 GPT-6 Luna 26.5%
- 8
Gemini 3.8 Flash ↑1 25%
- 9 GPT-6 Astra ↓1 24.5%
- 10
Gemini 4 Hydrogen 18.9%
The best score, 58.1%, costs $11 for the full run
Runs on the cost frontier score higher than any run that's cheaper. The costs are the leaderboard's own estimates, so compare them only within this benchmark.
Source: Official Construction's Last Exam leaderboard
| Model | Effort | Harness | Score | Cost of the full run | Output tokens |
|---|---|---|---|---|---|
| GPT-6.2 Astra | 58.1% | $11 | |||
| GPT-6.2 Sol | 44.2% | $2.80 | |||
| GPT-6.1 Terra | 36.4% | $1.40 | |||
| GPT-6 Sol | 34% | $3.50 | |||
| 28.3% | $4.20 | ||||
| 25.7% | $2.80 | ||||
| GPT-6 Luna | 21.6% | $0.20 | |||
| GPT-6 Astra | 19.2% | $14 | |||
| 19.1% | $1.20 | ||||
| 16.7% | $2.10 | ||||
| 16.6% | $1.10 | ||||
| Claude Opus 5.5 | 16.2% | $8.00 | |||
| Claude Opus 5 | 12.7% | $7.50 | |||
| Claude Fable 5.1 | 12.5% | $13 | |||
| 12% | $1.00 | ||||
| GPT-5.6 Sol | 11.1% | $5.50 | |||
| GPT-5.6 Luna | 10.1% | $0.30 | |||
| 6.00% | $0.40 | ||||
| GLM-5.3-Flash | 5.50% | $0.20 | |||
| 4.90% | $0.60 | ||||
| 4.00% | $0.20 | ||||
| 3.80% | $0.50 | ||||
| Claude Opus 4.6 | 3.40% | $6.50 | |||
| Claude Haiku 4.5 | 3.30% | $1.30 | |||
| Claude Sonnet 4.5 | 2.20% | $4.00 | |||
| 2.20% | $0.40 | ||||
| Claude Sonnet 5 | 2.10% | $3.00 | |||
| 1.80% | $0.30 | ||||
| GLM-4.6V | 1.30% | $0.40 | |||
| 1.30% | $2.20 |
The best score, 58.1%, costs $11 for the full run
Runs on the cost frontier score higher than any run that's cheaper. The costs are the leaderboard's own estimates, so compare them only within this benchmark.
Source: Official Construction's Last Exam leaderboard
| Model | Effort | Harness | Score | Cost of the full run | Output tokens |
|---|---|---|---|---|---|
| GPT-6.2 Astra | 58.1% | $11 | |||
| GPT-6.2 Sol | 44.2% | $2.80 | |||
| GPT-6.1 Terra | 36.4% | $1.40 | |||
| GPT-6 Sol | 34% | $3.50 | |||
| 28.3% | $4.20 | ||||
| 25.7% | $2.80 | ||||
| GPT-6 Luna | 21.6% | $0.20 | |||
| GPT-6 Astra | 19.2% | $14 | |||
| 19.1% | $1.20 | ||||
| 16.7% | $2.10 | ||||
| 16.6% | $1.10 | ||||
| Claude Opus 5.5 | 16.2% | $8.00 | |||
| Claude Opus 5 | 12.7% | $7.50 | |||
| Claude Fable 5.1 | 12.5% | $13 | |||
| 12% | $1.00 | ||||
| GPT-5.6 Sol | 11.1% | $5.50 | |||
| GPT-5.6 Luna | 10.1% | $0.30 | |||
| 6.00% | $0.40 | ||||
| GLM-5.3-Flash | 5.50% | $0.20 | |||
| 4.90% | $0.60 | ||||
| 4.00% | $0.20 | ||||
| 3.80% | $0.50 | ||||
| Claude Opus 4.6 | 3.40% | $6.50 | |||
| Claude Haiku 4.5 | 3.30% | $1.30 | |||
| Claude Sonnet 4.5 | 2.20% | $4.00 | |||
| 2.20% | $0.40 | ||||
| Claude Sonnet 5 | 2.10% | $3.00 | |||
| 1.80% | $0.30 | ||||
| GLM-4.6V | 1.30% | $0.40 | |||
| 1.30% | $2.20 |
3 models have held number one on the official leaderboard since its release in October 2026
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official Construction's Last Exam 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 Construction's Last Exam 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.
Versions
We switch the top row to each new version when it comes out. Open it for one row per version. An old version stays in play while people still test it, so whoever first takes it past 85% gets credit for closing it out.
Source: Official leaderboards
More in Economic work
- JobBench In use 130 tasks from 35 occupations workers most want delegated. Agent works through messy reference files, graded by fact-anchored binary rubrics. GLM-5.3 61.4%
- GDPval In use Real work tasks from 44 occupations with reference files. Experts blindly compare model deliverables with expert work. Metric is win-or-tie rate.
Seed2.1 Pro 87.9%
- GDPval-AA v2.1 In use Agents with shell and web access complete 220 GDPval gold tasks from 44 occupations. Blind pairwise comparisons give Elo ratings. Claude Opus 5.5 1,846
- AA-Briefcase v1.1 In use Version 1.1 of Artificial Analysis's multi-hour office work evaluation, with Elo ratings fitted by Crowd-BT. Claude Opus 5.5 1,822
- Agents' Last Exam v1.1 In use Agents complete 122 public long-horizon professional workflow tasks across 55 sub-industries. Deterministic graders check outcomes against hidden references. GPT-6.2 Astra 47.5%
- CoWorkBench In use Agent completes long-horizon productivity tasks across computer science, finance, law and medicine. Qwen in-house scoring (method unpublished). Claude Fable 5 75.9%
Data audit
All 11 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 5 pass
- Pass Independent runs Scores from the official leaderboard (32 entries).
- Pass Models scored 30 models we track, plus 2 other entries on its leaderboard.
- Pass Newest score The newest is from November 2026, for Gemini 4 Neon.
- Pass Frontier models 6 of the top 10 have a score, led by GPT-6.2 Astra (#1).
- Pass Headroom The best is 58.1%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on cle.togal.ai: 30 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.