AnalogCoder-Pro
Agents write analog circuit programs and use simulated waveforms to diagnose failures. The authors report the fraction of successful designs and the number of tasks solved.
ChipOmni ran three models on the journal paper’s 30 circuits. The preprint reports 28 tasks. The current task file lists 29 design tasks.
- Made by
- Yao Lai and colleagues
- Kind
- Electronics
- Level
- Expert
- Scored
- Checked automatically
- Answers
- Agent task
- Tasks
- All public
- Measures
- Capability
- Language
- English
- In and out
- Text, Code, Time series → Code, Actions
- Interaction
- Agentic
- Environment
- A simulator
- Refresh
- One fixed set
- Run by
- Anyone
- Based on
- AnalogCoder
- Strongest model tested
- GPT-5 #71 on the capability index
- Last model added
- September 2025
- Scores refreshed
- November 20th, 2026
Claude 3.7 Sonnet leads with 88.4% among models released by September 2025 Claude 3.7 Sonnet leads with 88.4% among models released by September 2025
Best score per entry 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. 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.
The newest model with a score here came out in September 2025, 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 Claude 3.7 Sonnet 88.4%
- 2 GPT-5 87.9%
- 3
Gemini 2.5 Flash Sep 2025 78.3%
- 4 DeepSeek-V3 (snapshot unspecified) 77.6%
- 5 GPT-4.1 76.4%
- 6 GPT-4.1 nano 36.7%
- 7
Qwen2.5-Coder-32B 33.1%
Source: The AnalogCoder-Pro paper
- 1 Claude 3.7 Sonnet 88.4%
- 2 GPT-5 87.9%
- 3
Gemini 2.5 Flash Sep 2025 78.3%
- 4 GPT-4.1 76.4%
- 5 GPT-4.1 nano 36.7%
- 6
Qwen2.5-Coder-32B 33.1%
Source: The AnalogCoder-Pro paper
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Claude 3.7 Sonnet | 88.4% | AnalogCoder-Pro | Feb 2025 | Official leaderboard |
| 2 | GPT-5 | 87.9% | AnalogCoder-Pro | Aug 2025 | Official leaderboard |
| 3 | | 78.3% | AnalogCoder-Pro | Sep 2025 | Official leaderboard |
| 4 | DeepSeek-V3 (snapshot unspecified) | 77.6% | AnalogCoder-Pro | Official leaderboard | |
| 5 | GPT-4.1 | 76.4% | AnalogCoder-Pro | Apr 2025 | Official leaderboard |
| 6 | GPT-4.1 nano | 36.7% | AnalogCoder-Pro | Apr 2025 | Official leaderboard |
| 7 | | 33.1% | AnalogCoder-Pro | Sep 2024 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Claude 3.7 Sonnet | 88.4% | AnalogCoder-Pro | Feb 2025 | Official leaderboard |
| 2 | GPT-5 | 87.9% | AnalogCoder-Pro | Aug 2025 | Official leaderboard |
| 3 | | 78.3% | AnalogCoder-Pro | Sep 2025 | Official leaderboard |
| 4 | GPT-4.1 | 76.4% | AnalogCoder-Pro | Apr 2025 | Official leaderboard |
| 5 | GPT-4.1 nano | 36.7% | AnalogCoder-Pro | Apr 2025 | Official leaderboard |
| 6 | | 33.1% | AnalogCoder-Pro | Sep 2024 | Official leaderboard |
Rankings by part
We read 30 parts from the benchmark’s paper, each ranking models on one part of the test, such as Task 1: Amplifier, Task 2: Amplifier and Task 3: Amplifier. Claude 3.7 Sonnet leads 21 of them, and 4 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.
Task 1: Amplifier
6 models we track have a score here.
- 1 Claude 3.7 Sonnet 100%
- 2 GPT-4.1 ↑2 100%
- 3 GPT-5 ↓1 100%
- 4
Gemini 2.5 Flash Sep 2025 ↓1 100%
- 5
Qwen2.5-Coder-32B ↑1 80%
- 6 GPT-4.1 nano ↓1 73.3%
Task 2: Amplifier
6 models we track have a score here.
- 1 Claude 3.7 Sonnet 100%
- 2 GPT-4.1 ↑2 100%
- 3 GPT-5 ↓1 100%
- 4
Gemini 2.5 Flash Sep 2025 ↓1 100%
- 5
Qwen2.5-Coder-32B ↑1 80%
- 6 GPT-4.1 nano ↓1 6.70%
Task 3: Amplifier
6 models we track have a score here.
- 1 Claude 3.7 Sonnet 100%
- 2 GPT-4.1 ↑2 100%
- 3 GPT-5 ↓1 100%
- 4
Gemini 2.5 Flash Sep 2025 ↓1 100%
- 5
Qwen2.5-Coder-32B ↑1 70%
- 6 GPT-4.1 nano ↓1 33.3%
Task 4: Amplifier
6 models we track have a score here.
- 1 Claude 3.7 Sonnet 100%
- 2 GPT-4.1 ↑2 100%
- 3 GPT-5 ↓1 100%
- 4 GPT-4.1 nano ↑1 90%
- 5
Gemini 2.5 Flash Sep 2025 ↓2 80%
- 6
Qwen2.5-Coder-32B 66.7%
Task 5: Amplifier
6 models we track have a score here.
- 1 Claude 3.7 Sonnet 100%
- 2 GPT-4.1 ↑2 100%
- 3 GPT-5 ↓1 100%
- 4
Gemini 2.5 Flash Sep 2025 ↓1 93.3%
- 5 GPT-4.1 nano 26.7%
- 6
Qwen2.5-Coder-32B 20%
Task 6: Inverter
6 models we track have a score here.
- 1
Qwen2.5-Coder-32B ↑5 100%
- 2 Claude 3.7 Sonnet ↓1 100%
- 3 GPT-4.1 ↑1 100%
- 4 GPT-5 ↓2 100%
- 5
Gemini 2.5 Flash Sep 2025 ↓2 100%
- 6 GPT-4.1 nano ↓1 86.7%
Task 7: Inverter
6 models we track have a score here.
- 1
Qwen2.5-Coder-32B ↑5 100%
- 2 Claude 3.7 Sonnet ↓1 100%
- 3 GPT-4.1 ↑1 100%
- 4 GPT-5 ↓2 100%
- 5
Gemini 2.5 Flash Sep 2025 ↓2 100%
- 6 GPT-4.1 nano ↓1 86.7%
Task 8: Current Mirror
6 models we track have a score here.
- 1 Claude 3.7 Sonnet 100%
- 2 GPT-4.1 ↑2 100%
- 3 GPT-5 ↓1 100%
- 4
Gemini 2.5 Flash Sep 2025 ↓1 76.7%
- 5
Qwen2.5-Coder-32B ↑1 73.3%
- 6 GPT-4.1 nano ↓1 30%
Task 9: Comparator
6 models we track have a score here.
- 1 Claude 3.7 Sonnet 100%
- 2 GPT-4.1 ↑2 100%
- 3 GPT-4.1 nano ↑2 73.3%
- 4
Qwen2.5-Coder-32B ↑2 53.3%
- 5 GPT-5 ↓3 53.3%
- 6
Gemini 2.5 Flash Sep 2025 ↓3 3.30%
Task 10: Filter
6 models we track have a score here.
- 1 Claude 3.7 Sonnet 100%
- 2 GPT-5 100%
- 3 GPT-4.1 ↑1 96.7%
- 4
Gemini 2.5 Flash Sep 2025 ↓1 96.7%
- 5 GPT-4.1 nano 93.3%
- 6
Qwen2.5-Coder-32B 80%
Task 11: Filter
6 models we track have a score here.
- 1 Claude 3.7 Sonnet 100%
- 2 GPT-5 100%
- 3
Gemini 2.5 Flash Sep 2025 100%
- 4 GPT-4.1 90%
- 5 GPT-4.1 nano 86.7%
- 6
Qwen2.5-Coder-32B 13.3%
Task 12: Filter
6 models we track have a score here.
- 1 Claude 3.7 Sonnet 100%
- 2
Gemini 2.5 Flash Sep 2025 ↑1 100%
- 3 GPT-5 ↓1 96.7%
- 4 GPT-4.1 80%
- 5
Qwen2.5-Coder-32B ↑1 16.7%
- 6 GPT-4.1 nano ↓1 13.3%
Task 13: Filter
6 models we track have a score here.
- 1 GPT-5 ↑1 100%
- 2
Gemini 2.5 Flash Sep 2025 ↑1 100%
- 3 Claude 3.7 Sonnet ↓2 66.7%
- 4 GPT-4.1 36.7%
- 5 GPT-4.1 nano 6.70%
- 6
Qwen2.5-Coder-32B 0%
Task 14: Amplifier
6 models we track have a score here.
- 1 Claude 3.7 Sonnet 100%
- 2 GPT-4.1 ↑2 100%
- 3 GPT-5 ↓1 100%
- 4
Qwen2.5-Coder-32B ↑2 43.3%
- 5 GPT-4.1 nano 36.7%
- 6
Gemini 2.5 Flash Sep 2025 ↓3 26.7%
Task 15: Amplifier
6 models we track have a score here.
- 1 Claude 3.7 Sonnet 100%
- 2 GPT-4.1 ↑2 100%
- 3 GPT-5 ↓1 100%
- 4
Gemini 2.5 Flash Sep 2025 ↓1 100%
- 5 GPT-4.1 nano 66.7%
- 6
Qwen2.5-Coder-32B 30%
Task 16: Opamp
6 models we track have a score here.
- 1 Claude 3.7 Sonnet 100%
- 2
Gemini 2.5 Flash Sep 2025 ↑1 100%
- 3 GPT-5 ↓1 96.7%
- 4 GPT-4.1 76.7%
- 5 GPT-4.1 nano 16.7%
- 6
Qwen2.5-Coder-32B 0%
Task 17: Current Mirror
6 models we track have a score here.
- 1 GPT-4.1 ↑3 66.7%
- 2 GPT-5 66.7%
- 3
Gemini 2.5 Flash Sep 2025 30%
- 4 Claude 3.7 Sonnet ↓3 6.70%
- 5 GPT-4.1 nano 3.30%
- 6
Qwen2.5-Coder-32B 0%
Task 18: Opamp
6 models we track have a score here.
- 1 Claude 3.7 Sonnet 100%
- 2 GPT-5 100%
- 3
Gemini 2.5 Flash Sep 2025 96.7%
- 4 GPT-4.1 93.3%
- 5 GPT-4.1 nano 16.7%
- 6
Qwen2.5-Coder-32B 0%
Task 19: Mixer
6 models we track have a score here.
- 1 Claude 3.7 Sonnet 100%
- 2 GPT-5 96.7%
- 3 GPT-4.1 ↑1 50%
- 4
Qwen2.5-Coder-32B ↑2 20%
- 5 GPT-4.1 nano 6.70%
- 6
Gemini 2.5 Flash Sep 2025 ↓3 3.30%
Task 20: Opamp
6 models we track have a score here.
- 1 Claude 3.7 Sonnet 100%
- 2 GPT-5 100%
- 3
Gemini 2.5 Flash Sep 2025 100%
- 4 GPT-4.1 90%
- 5
Qwen2.5-Coder-32B ↑1 6.70%
- 6 GPT-4.1 nano ↓1 0%
Task 21: Opamp
6 models we track have a score here.
- 1
Gemini 2.5 Flash Sep 2025 ↑2 90%
- 2 Claude 3.7 Sonnet ↓1 86.7%
- 3 GPT-5 ↓1 56.7%
- 4 GPT-4.1 46.7%
- 5
Qwen2.5-Coder-32B ↑1 0%
- 6 GPT-4.1 nano ↓1 0%
Task 22: Oscillator
6 models we track have a score here.
- 1
Gemini 2.5 Flash Sep 2025 ↑2 66.7%
- 2 Claude 3.7 Sonnet ↓1 46.7%
- 3 GPT-4.1 ↑1 46.7%
- 4
Qwen2.5-Coder-32B ↑2 16.7%
- 5 GPT-4.1 nano 16.7%
- 6 GPT-5 ↓4 16.7%
Task 23: Oscillator
6 models we track have a score here.
- 1 Claude 3.7 Sonnet 100%
- 2 GPT-5 100%
- 3
Gemini 2.5 Flash Sep 2025 76.7%
- 4 GPT-4.1 73.3%
- 5
Qwen2.5-Coder-32B ↑1 0%
- 6 GPT-4.1 nano ↓1 0%
Task 24: Integrator
6 models we track have a score here.
- 1 Claude 3.7 Sonnet 100%
- 2 GPT-5 100%
- 3
Gemini 2.5 Flash Sep 2025 100%
- 4 GPT-4.1 96.7%
- 5 GPT-4.1 nano 80%
- 6
Qwen2.5-Coder-32B 13.3%
Task 25: Differentiator
6 models we track have a score here.
- 1 Claude 3.7 Sonnet 100%
- 2 GPT-4.1 ↑2 100%
- 3 GPT-5 ↓1 100%
- 4
Gemini 2.5 Flash Sep 2025 ↓1 100%
- 5 GPT-4.1 nano 40%
- 6
Qwen2.5-Coder-32B 30%
Task 26: Adder
6 models we track have a score here.
- 1 Claude 3.7 Sonnet 100%
- 2
Gemini 2.5 Flash Sep 2025 ↑1 100%
- 3 GPT-5 ↓1 80%
- 4 GPT-4.1 nano ↑1 63.3%
- 5
Qwen2.5-Coder-32B ↑1 60%
- 6 GPT-4.1 ↓2 30%
Task 27: Subtractor
6 models we track have a score here.
- 1 Claude 3.7 Sonnet 100%
- 2 GPT-5 96.7%
- 3
Gemini 2.5 Flash Sep 2025 93.3%
- 4 GPT-4.1 73.3%
- 5 GPT-4.1 nano 26.7%
- 6
Qwen2.5-Coder-32B 0%
Task 28: Schmitt trigger
6 models we track have a score here.
- 1
Gemini 2.5 Flash Sep 2025 ↑2 93.3%
- 2 GPT-5 90%
- 3 Claude 3.7 Sonnet ↓2 66.7%
- 4 GPT-4.1 33.3%
- 5
Qwen2.5-Coder-32B ↑1 20%
- 6 GPT-4.1 nano ↓1 20%
Task 29: VCO
6 models we track have a score here.
- 1 GPT-5 ↑1 86.7%
- 2 Claude 3.7 Sonnet ↓1 80%
- 3
Gemini 2.5 Flash Sep 2025 23.3%
- 4 GPT-4.1 13.3%
- 5
Qwen2.5-Coder-32B ↑1 0%
- 6 GPT-4.1 nano ↓1 0%
Task 30: PLL
6 models we track have a score here.
- 1
Qwen2.5-Coder-32B ↑5 0%
- 2 Claude 3.7 Sonnet ↓1 0%
- 3 GPT-4.1 ↑1 0%
- 4 GPT-4.1 nano ↑1 0%
- 5 GPT-5 ↓3 0%
- 6
Gemini 2.5 Flash Sep 2025 ↓3 0%
2 models have held number one in the benchmark’s paper since 2023
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: The AnalogCoder-Pro paper
Claude 3.7 Sonnet on the official leaderboard passed 85% in February 2025
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official AnalogCoder-Pro 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 8 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 2 pass 3 notes
- Pass Independent runs Scores from the benchmark’s paper (7 entries).
- Pass Models scored 6 models we track, plus 1 other entry on its leaderboard.
- Note Newest score The newest came 14 months ago, in September 2025, for Gemini 2.5 Flash. That’s expected for a saturated test.
- Note Frontier models None of the top 10 has a score, which is fine for a retired test. The strongest tested is GPT-5 (#71).
- Note Headroom Saturated. The best is 88.4%, past 85% since February 2025.
Leaderboard import 3 pass
- Pass Importer We read the board on doi.org: 6 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.