PLCWorld
Models write Structured Text programs for motion control and material handling. A plant simulator measures task completion and safety violations separately.
The authors compare six models and four adapted workflows on 100 tasks with 473 operating conditions.
- Made by
- Dongguk University and MOAI technologies
- Released
- October 2026
- Kind
- Electronics
- Level
- University
- Scored
- Checked automatically
- Answers
- Code
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Single turn
- Refresh
- One fixed set
- Run by
- Anyone
- Strongest model tested
- GPT-5.5 #21 on the capability index
- Last model added
- September 2026
- Scores refreshed
- November 20th, 2026
AutoPLC + GPT-5.5 tops the leaderboard with 63% GPT-5.5 leads with 62.6% among models released by September 2026
Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran on its own with several settings, we keep its best. Agents other companies built around a model get a pale shade of that model's lab color. Gray marks blends and entries not in our model list, such as fine-tunes and older models. None of these count toward leaders. The best model on its own is GPT-5.5, at 62.6%. Agents built around a model, such as AutoPLC + GPT-5.5, are their own entries. We don't count them as the model's score. This view shows only models we track, each at its best.
The newest model with a score here came out in September 2026, 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 AutoPLC + GPT-5.5 Matched workflow comparison 63%
- 2 GPT-5.5 Direct generation 62.6%
- 3
Gemini 3.8 Flash Direct generation 61.2%
- 4 LLM4PLC + GPT-5.5 Matched workflow comparison 57.9%
- 5 SemaPLC + GPT-5.5 Matched workflow comparison 55.4%
- 6
Gemini 3.1 Pro Direct generation 53.4%
- 7 Agents4PLC + GPT-5.5 Matched workflow comparison 52.9%
- 8 Claude Sonnet 5 Direct generation 49.2%
- 9
Qwen3.5-Plus Direct generation 38.3%
- 10 GPT-4o (snapshot unspecified) Direct generation 2.18%
Source: Official PLCWorld leaderboard
- 1 GPT-5.5 Direct generation 62.6%
- 2
Gemini 3.8 Flash Direct generation 61.2%
- 3
Gemini 3.1 Pro Direct generation 53.4%
- 4 Claude Sonnet 5 Direct generation 49.2%
- 5
Qwen3.5-Plus Direct generation 38.3%
Source: Official PLCWorld leaderboard
| # | Entry | Score ± | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | AutoPLC + GPT-5.5 | 63% ±8.5 | Matched workflow comparison | Apr 2026 | Official leaderboard |
| 2 | GPT-5.5 | 62.6% ±7.2 | Direct generation | Apr 2026 | Official leaderboard |
| 3 | | 61.2% ±5.4 | Direct generation | Sep 2026 | Official leaderboard |
| 4 | LLM4PLC + GPT-5.5 | 57.9% ±8.3 | Matched workflow comparison | Apr 2026 | Official leaderboard |
| 5 | SemaPLC + GPT-5.5 | 55.4% ±7.4 | Matched workflow comparison | Apr 2026 | Official leaderboard |
| 6 | | 53.4% ±5.9 | Direct generation | Feb 2026 | Official leaderboard |
| 7 | Agents4PLC + GPT-5.5 | 52.9% 44.8 to 61.3 | Matched workflow comparison | Apr 2026 | Official leaderboard |
| 8 | Claude Sonnet 5 | 49.2% ±6.1 | Direct generation | Jun 2026 | Official leaderboard |
| 9 | | 38.3% ±5.7 | Direct generation | Feb 2026 | Official leaderboard |
| 10 | GPT-4o (snapshot unspecified) | 2.18% 0.7 to 3.9 | Direct generation | Official leaderboard |
| # | Model | Score ± | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-5.5 | 62.6% ±7.2 | Direct generation | Apr 2026 | Official leaderboard |
| 2 | | 61.2% ±5.4 | Direct generation | Sep 2026 | Official leaderboard |
| 3 | | 53.4% ±5.9 | Direct generation | Feb 2026 | Official leaderboard |
| 4 | Claude Sonnet 5 | 49.2% ±6.1 | Direct generation | Jun 2026 | Official leaderboard |
| 5 | | 38.3% ±5.7 | Direct generation | Feb 2026 | Official leaderboard |
Rankings by part
We read 6 parts from the benchmark’s own leaderboard, each ranking models on one part of the test, such as Motion Control: Easy, Motion Control: Medium and Motion Control: Hard. Gemini 3.8 Flash leads 3 of them, and GPT-5.5 leads the other 3. Arrows show how far a model moved from its overall rank among the models we track.
Motion Control: Easy
5 models we track have a score here.
- 1
Gemini 3.8 Flash ↑1 95.6%
- 2 Claude Sonnet 5 ↑2 77.8%
- 3
Gemini 3.1 Pro 73.3%
- 4 GPT-5.5 ↓3 73.3%
- 5
Qwen3.5-Plus 65.6%
Motion Control: Medium
5 models we track have a score here.
- 1
Gemini 3.8 Flash ↑1 76.9%
- 2 GPT-5.5 ↓1 75.6%
- 3 Claude Sonnet 5 ↑1 72.3%
- 4
Gemini 3.1 Pro ↓1 63%
- 5
Qwen3.5-Plus 55.1%
Motion Control: Hard
5 models we track have a score here.
- 1 GPT-5.5 22.9%
- 2
Gemini 3.1 Pro ↑1 11.8%
- 3
Gemini 3.8 Flash ↓1 4.86%
- 4
Qwen3.5-Plus ↑1 0%
- 5 Claude Sonnet 5 ↓1 0%
Material Handling: Easy
5 models we track have a score here.
- 1
Gemini 3.8 Flash ↑1 100%
- 2 GPT-5.5 ↓1 97.8%
- 3
Gemini 3.1 Pro 88.4%
- 4
Qwen3.5-Plus ↑1 86.2%
- 5 Claude Sonnet 5 ↓1 85.5%
Material Handling: Medium
5 models we track have a score here.
- 1 GPT-5.5 93.1%
- 2
Gemini 3.8 Flash 88.5%
- 3
Gemini 3.1 Pro 74.7%
- 4 Claude Sonnet 5 71.2%
- 5
Qwen3.5-Plus 46.2%
Material Handling: Hard
5 models we track have a score here.
- 1 GPT-5.5 28.7%
- 2
Gemini 3.1 Pro ↑1 22.1%
- 3
Gemini 3.8 Flash ↓1 17.2%
- 4 Claude Sonnet 5 4.37%
- 5
Qwen3.5-Plus 1.64%
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official PLCWorld 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 PLCWorld 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 All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 4 pass 1 note
- Pass Independent runs Scores from the official leaderboard (10 entries).
- Pass Models scored 5 models we track, plus 5 other entries on its leaderboard.
- Pass Newest score The newest is from September 2026, for Gemini 3.8 Flash.
- Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5.5 (#21).
- Pass Headroom The best is 62.6%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on yunji0516.github.io: 10 rows for models we track and 1 other entry 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.