SemaPLC Function Track
Agents generate individual Structured Text program units. A compiler and a formal verifier check the program against the requirements. Inconclusive checks count as failures.
Engineers revised 53 of 117 Agents4PLC tasks. The resulting set has 607 verification properties. The public download includes the original and revised data.
- Made by
- Midea AIRC and collaborators
- Released
- August 2026
- Kind
- Electronics
- Level
- Expert
- Scored
- Checked automatically
- Answers
- Code
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Single turn
- Items from
- Another benchmark, Written by experts
- Refresh
- One fixed set
- Run by
- Anyone
- Based on
- A version of another benchmark
- Strongest model tested
- GPT-5.5 #21 on the capability index
- Last model added
- June 2026
- Scores refreshed
- November 20th, 2026
GPT-5.5 leads with 82.1% among models released by June 2026
Best score per model from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart.
The newest model with a score here came out in June 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 GPT-5.5 full 82.1%
- 2 GLM-5.2 full 76.1%
- 3
Qwen3.5-Plus full 75.2%
- 4
MiniMax-M2.7 full 69.2%
- 5
DeepSeek-V4-Pro full 69.2%
- 6
MiniMax-M3 full 69.2%
- 7
DeepSeek-V4-Flash full 67.5%
Source: The SemaPLC Function Track paper
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-5.5 | 82.1% | full, SemaPLC | Apr 2026 | Official leaderboard |
| 2 | GLM-5.2 | 76.1% | full, SemaPLC | Jun 2026 | Official leaderboard |
| 3 | | 75.2% | full, SemaPLC | Feb 2026 | Official leaderboard |
| 4 | | 69.2% | full, SemaPLC | Mar 2026 | Official leaderboard |
| 5 | | 69.2% | full, SemaPLC | Apr 2026 | Official leaderboard |
| 6 | | 69.2% | full, SemaPLC | Jun 2026 | Official leaderboard |
| 7 | | 67.5% | full, SemaPLC | Apr 2026 | Official leaderboard |
Scores at each effort level
Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.
Source: The SemaPLC Function Track paper
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| GPT-5.5 | full | SemaPLC | 82.1% | ||
| GPT-5.5 | AutoPLC | 79.5% | |||
| GPT-5.5 | Agents4PLC | 78.6% | |||
| GLM-5.2 | full | SemaPLC | 76.1% | ||
| full | SemaPLC | 75.2% | |||
| GLM-5.2 | Agents4PLC | 74.4% | |||
| GPT-5.5 | bare | SemaPLC | 71.8% | ||
| full | SemaPLC | 69.2% | |||
| full | SemaPLC | 69.2% | |||
| full | SemaPLC | 69.2% | |||
| AutoPLC | 67.5% | ||||
| Agents4PLC | 67.5% | ||||
| full | SemaPLC | 67.5% | |||
| AutoPLC | 65% | ||||
| GLM-5.2 | bare | SemaPLC | 63.2% | ||
| bare | SemaPLC | 62.4% | |||
| Agents4PLC | 62.4% | ||||
| AutoPLC | 61.5% | ||||
| bare | SemaPLC | 60.7% | |||
| GLM-5.2 | AutoPLC | 59% | |||
| bare | SemaPLC | 55.6% | |||
| Agents4PLC | 55.6% | ||||
| AutoPLC | 54.7% | ||||
| Agents4PLC | 54.7% | ||||
| Agents4PLC | 53.8% | ||||
| AutoPLC | 49.6% | ||||
| GPT-5.5 | LLM4PLC | 44.4% | |||
| LLM4PLC | 43.6% | ||||
| LLM4PLC | 41% | ||||
| bare | SemaPLC | 39.3% | |||
| bare | SemaPLC | 34.2% | |||
| GLM-5.2 | LLM4PLC | 30.8% | |||
| LLM4PLC | 22.2% | ||||
| LLM4PLC | 15.4% | ||||
| LLM4PLC | 13.7% |
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: The SemaPLC Function Track paper
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official SemaPLC Function Track 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 benchmark’s paper (7 entries).
- Pass Models scored 7 models we track.
- Pass Newest score The newest is from June 2026, for GLM-5.2.
- Note Frontier models None of the top 10 has a score. The strongest tested is GPT-5.5 (#21).
- Pass Headroom The best is 82.1%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on arxiv.org: 35 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.