ICLayout-Bench
Agents turn transistor netlists into chip layouts. Design tools check the layout rules and connectivity, then measure electrical performance and area.
The makers compare seven models at 22 model and effort settings. They score 21 public circuits and 19 private circuits separately.
- Made by
- Zhangmai Li and colleagues
- Kind
- Electronics
- Level
- Expert
- Scored
- Checked automatically
- Answers
- Agent task
- Tasks
- Public, plus a private test set
- Measures
- Capability
- Language
- English
- Interaction
- Agentic
- Environment
- Terminal, A simulator
- 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
GPT-6 Astra leads with 80.3 GPT-6 Astra leads with 80.3
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. This view shows only models we track, each at its best.
- 1 GPT-6 Astra medium 80.3
- 2 GPT-6.1 Sol xhigh 73.8
- 3 GPT-6.1 Terra xhigh 60.6
- 4 k3.1-pro max 43.6
- 5 deepseek-flash high 33.4
- 6 GPT-6 Luna xhigh 29.3
- 7 GLM-5.3-Flash max 4.52
Source: Official ICLayout-Bench leaderboard
- 1 GPT-6 Astra medium 80.3
- 2 GPT-6.1 Sol xhigh 73.8
- 3 GPT-6.1 Terra xhigh 60.6
- 4 GPT-6 Luna xhigh 29.3
- 5 GLM-5.3-Flash max 4.52
Source: Official ICLayout-Bench leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-6 Astra | 80.3 | medium, codex | Sep 2026 | Official leaderboard |
| 2 | GPT-6.1 Sol | 73.8 | xhigh, codex | Sep 2026 | Official leaderboard |
| 3 | GPT-6.1 Terra | 60.6 | xhigh, codex | Oct 2026 | Official leaderboard |
| 4 | k3.1-pro | 43.6 | max, kimi-code | Official leaderboard | |
| 5 | deepseek-flash | 33.4 | high, dsh | Official leaderboard | |
| 6 | GPT-6 Luna | 29.3 | xhigh, codex | Sep 2026 | Official leaderboard |
| 7 | GLM-5.3-Flash | 4.52 | max, zcode | Aug 2026 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-6 Astra | 80.3 | medium, codex | Sep 2026 | Official leaderboard |
| 2 | GPT-6.1 Sol | 73.8 | xhigh, codex | Sep 2026 | Official leaderboard |
| 3 | GPT-6.1 Terra | 60.6 | xhigh, codex | Oct 2026 | Official leaderboard |
| 4 | GPT-6 Luna | 29.3 | xhigh, codex | Sep 2026 | Official leaderboard |
| 5 | GLM-5.3-Flash | 4.52 | max, zcode | Aug 2026 | Official leaderboard |
Rankings by part
We read 2 parts from the benchmark’s own leaderboard: Public and Private. Each ranks models on one part of the test. GPT-6 Astra leads Public and GPT-6.1 Sol leads Private. Arrows show how far a model moved from its overall rank among the models we track.
Public
5 models we track have a score here.
- 1 GPT-6 Astra 80.3
- 2 GPT-6.1 Sol 73.8
- 3 GPT-6.1 Terra 60.6
- 4 GPT-6 Luna 29.3
- 5 GLM-5.3-Flash 4.52
Private
5 models we track have a score here.
- 1 GPT-6.1 Sol ↑1 67.0
- 2 GPT-6 Astra ↓1 66.9
- 3 GPT-6.1 Terra 52.8
- 4 GPT-6 Luna 12.3
- 5 GLM-5.3-Flash 2.02
The best score, 80.3, costs $25 a task
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 ICLayout-Bench leaderboard
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| GPT-6 Astra | medium | codex | 80.3 | $25 | 99,253 |
| GPT-6 Astra | xhigh | codex | 78.2 | $22 | 99,751 |
| GPT-6 Astra | high | codex | 77.2 | $24 | 105,029 |
| GPT-6 Astra | low | codex | 77.1 | $25 | 89,609 |
| GPT-6.1 Sol | xhigh | codex | 73.8 | $4.82 | 131,208 |
| GPT-6.1 Sol | medium | codex | 72.6 | $4.77 | 115,896 |
| GPT-6.1 Sol | low | codex | 70.1 | $4.63 | 112,284 |
| GPT-6.1 Sol | high | codex | 69.8 | $5.20 | 143,026 |
| GPT-6.1 Terra | xhigh | codex | 60.6 | $5.65 | 107,594 |
| GPT-6.1 Terra | high | codex | 57.1 | $3.40 | 61,720 |
| GPT-6.1 Terra | low | codex | 52.0 | $2.19 | 43,136 |
| GPT-6.1 Terra | medium | codex | 48.5 | $2.69 | 54,887 |
| GPT-6 Luna | xhigh | codex | 29.3 | $0.77 | 420,192 |
| GPT-6 Luna | max | codex | 18.2 | $0.85 | 455,228 |
| GPT-6 Luna | high | codex | 13.6 | $0.42 | 164,142 |
| GPT-6 Luna | medium | codex | 5.18 | $0.22 | 81,664 |
| GLM-5.3-Flash | max | zcode | 4.52 | $0.63 | 212,224 |
| GLM-5.3-Flash | high | zcode | 4.22 | $0.60 | 209,610 |
The best score, 80.3, costs $25 a task
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 ICLayout-Bench leaderboard
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| GPT-6 Astra | medium | codex | 80.3 | $25 | 99,253 |
| GPT-6 Astra | xhigh | codex | 78.2 | $22 | 99,751 |
| GPT-6 Astra | high | codex | 77.2 | $24 | 105,029 |
| GPT-6 Astra | low | codex | 77.1 | $25 | 89,609 |
| GPT-6.1 Sol | xhigh | codex | 73.8 | $4.82 | 131,208 |
| GPT-6.1 Sol | medium | codex | 72.6 | $4.77 | 115,896 |
| GPT-6.1 Sol | low | codex | 70.1 | $4.63 | 112,284 |
| GPT-6.1 Sol | high | codex | 69.8 | $5.20 | 143,026 |
| GPT-6.1 Terra | xhigh | codex | 60.6 | $5.65 | 107,594 |
| GPT-6.1 Terra | high | codex | 57.1 | $3.40 | 61,720 |
| GPT-6.1 Terra | low | codex | 52.0 | $2.19 | 43,136 |
| GPT-6.1 Terra | medium | codex | 48.5 | $2.69 | 54,887 |
| GPT-6 Luna | xhigh | codex | 29.3 | $0.77 | 420,192 |
| GPT-6 Luna | max | codex | 18.2 | $0.85 | 455,228 |
| GPT-6 Luna | high | codex | 13.6 | $0.42 | 164,142 |
| GPT-6 Luna | medium | codex | 5.18 | $0.22 | 81,664 |
| GLM-5.3-Flash | max | zcode | 4.52 | $0.63 | 212,224 |
| GLM-5.3-Flash | high | zcode | 4.22 | $0.60 | 209,610 |
2 models have held number one on the official leaderboard since 2023
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official ICLayout-Bench 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 ICLayout-Bench 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 Engineering and CAD
- EEBench In use 13 electrical-engineering circuit and PCB design tasks graded by deterministic build checks, simulation and BOM constraints. No LLM judge.
Gemini 4 Carbon 83.9%
- 3DCodeBench In use Agentic procedural 3D modeling via code: author engine-ready 3D assets through software APIs. Graded for executability and shape fidelity.
LongCat-2.5 55.4%
- CADGenBench Only in system cards Agents construct CAD models from design prompts. The generation split is graded for executability and geometric correctness. Claude Fable 5 45.1%
- PartBench Only in system cards Parametric mechanical part design and CAD-style generation, graded for geometric and manufacturing fidelity. Grok 4.6 58.5%
- Razavi-Bench Not in model cards Models explain what happens in analog circuits and why. A language model awards up to four points per answer against a reference and a rubric.
Gemini 4 Carbon 98.3%
- CircuitRubric Not in model cards Models write analog circuit netlists from design requests. A program compares transistor connections and relative device sizes with accepted reference circuits. Claude Opus 4.8 96%
Data audit
All 10 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access Public tasks, plus a private test set.
- Pass Confidence Checked against a primary source.
Scores 4 pass 1 note
- Pass Independent runs Scores from the official leaderboard (7 entries).
- Pass Models scored 5 models we track, plus 2 other entries on its leaderboard.
- Pass Newest score The newest is from October 2026, for GPT-6.1 Terra.
- Pass Frontier models 3 of the top 10 have a score, led by GPT-6 Astra (#4).
- Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on lizhangmai.github.io: 18 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.