MMCircuitEval
Models answer questions about circuit diagrams, specifications, and design decisions. The authors combine text similarity with a language model’s judgment.
OpenReward ranks 33 published model results. ChipOmni separately tested three examples with Qwen.
- Made by
- Chenchen Zhao and colleagues
- Released
- July 2025
- Kind
- Electronics
- Level
- University
- Scored
- Graded by a model
- Answers
- Open-ended
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Single turn
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Vision
- Strongest model tested
- GPT-4 Turbo Apr 2024 #172 on the capability index
- Last model added
- March 2026
- Scores refreshed
- November 20th, 2026
GPT-4V leads with 69.4% among models released by March 2026 GPT-4V leads with 69.4% among models released by March 2026
Best score per entry from OpenReward’s leaderboard, read on November 20th, 2026. 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 March 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
- 69.4 GPT-4V
- 68.0 GPT-4o
- 67.4 GPT-4 Turbo
- 67.1 ChipExpert*
- 63.7 GPT-4
- 58.5 LlaMa3.2-Vision-Instruct-90B
- 55.7 Reka Flash
- 54.9 GPT-3.5 Turbo
- 53.0 QWen-2.5-Instruct-7B
- 50.2 QWen-2-Instruct-72B
- 48.2 QWen-2-Instruct-7B
- 43.2 LlaMa3.2-Vision-Instruct-11B
- 43.0
MiniCPM-Llama3-V 2.5
- 41.7
InternVL2-40B
- 40.5
DeepSeekMath-7B-Instruct
- 39.9 DeepSeek-Math-7B-RL
- 39.7 DeepSeek-V2-Lite-Chat
- 38.8
InternVL2-8B
- 36.7 Reka Edge
- 35.1 DeepSeek-LLM-7B-Chat
- 34.0 DeepSeek-MoE-16B-Chat
- 32.5
Yi-VL-34B
- 32.2 Qwen-VL-Chat
- 21.1
MiniCPM-V
- 19.7 InternLM-XComposer-VL-7B
- 17.3
Yi-VL-6B
- 14.3
BLIP-2 FLAN-T5-XL
- 14.0
BLIP-2 FLAN-T5-XXL
- 13.8 QWen-2-Instruct-0.5B
- 13.6
Kosmos-2
- 11.5
MiniCPM-V 2.0
- 10.8
InstructBLIP FLAN-T5-XL
- 10.7
InstructBLIP FLAN-T5-XXL
- 1 GPT-4V 69.4%
- 2 GPT-4o Mar 2025 68%
- 3 GPT-4 Turbo Apr 2024 67.4%
- 4 ChipExpert* 67.1%
- 5 GPT-4 Jun 2023 63.7%
- 6 LlaMa3.2-Vision-Instruct-90B 58.5%
- 7 Reka Flash 55.7%
- 8 GPT-3.5 Turbo Jan 2024 54.9%
- 9 QWen-2.5-Instruct-7B 53%
- 10 QWen-2-Instruct-72B 50.2%
- 11 QWen-2-Instruct-7B 48.2%
- 12 LlaMa3.2-Vision-Instruct-11B 43.2%
- 13
MiniCPM-Llama3-V 2.5 43%
- 14
InternVL2-40B 41.7%
- 15
DeepSeekMath-7B-Instruct 40.5%
- 16 DeepSeek-Math-7B-RL 39.9%
- 17 DeepSeek-V2-Lite-Chat 39.7%
- 18
InternVL2-8B 38.8%
- 19 Reka Edge 36.7%
- 20 DeepSeek-LLM-7B-Chat 35.1%
- 21 DeepSeek-MoE-16B-Chat 34%
- 22
Yi-VL-34B 32.5%
- 23 Qwen-VL-Chat 32.2%
- 24
MiniCPM-V 21.1%
- 25 InternLM-XComposer-VL-7B 19.7%
- 26
Yi-VL-6B 17.3%
- 27
BLIP-2 FLAN-T5-XL 14.3%
- 28
BLIP-2 FLAN-T5-XXL 14%
- 29 QWen-2-Instruct-0.5B 13.8%
- 30
Kosmos-2 13.6%
- 31
MiniCPM-V 2.0 11.5%
- 32
InstructBLIP FLAN-T5-XL 10.8%
- 33
InstructBLIP FLAN-T5-XXL 10.7%
Source: OpenReward’s leaderboard
- 69.4 GPT-4V
- 68.0 GPT-4o
- 67.4 GPT-4 Turbo
- 63.7 GPT-4
- 55.7 Reka Flash
- 54.9 GPT-3.5 Turbo
- 43.0
MiniCPM-Llama3-V 2.5
- 41.7
InternVL2-40B
- 40.5
DeepSeekMath-7B-Instruct
- 38.8
InternVL2-8B
- 36.7 Reka Edge
- 32.5
Yi-VL-34B
- 32.2 Qwen-VL-Chat
- 21.1
MiniCPM-V
- 17.3
Yi-VL-6B
- 14.3
BLIP-2 FLAN-T5-XL
- 14.0
BLIP-2 FLAN-T5-XXL
- 13.6
Kosmos-2
- 11.5
MiniCPM-V 2.0
- 10.8
InstructBLIP FLAN-T5-XL
- 10.7
InstructBLIP FLAN-T5-XXL
- 1 GPT-4V 69.4%
- 2 GPT-4o Mar 2025 68%
- 3 GPT-4 Turbo Apr 2024 67.4%
- 4 GPT-4 Jun 2023 63.7%
- 5 Reka Flash 55.7%
- 6 GPT-3.5 Turbo Jan 2024 54.9%
- 7
MiniCPM-Llama3-V 2.5 43%
- 8
InternVL2-40B 41.7%
- 9
DeepSeekMath-7B-Instruct 40.5%
- 10
InternVL2-8B 38.8%
- 11 Reka Edge 36.7%
- 12
Yi-VL-34B 32.5%
- 13 Qwen-VL-Chat 32.2%
- 14
MiniCPM-V 21.1%
- 15
Yi-VL-6B 17.3%
- 16
BLIP-2 FLAN-T5-XL 14.3%
- 17
BLIP-2 FLAN-T5-XXL 14%
- 18
Kosmos-2 13.6%
- 19
MiniCPM-V 2.0 11.5%
- 20
InstructBLIP FLAN-T5-XL 10.8%
- 21
InstructBLIP FLAN-T5-XXL 10.7%
Source: OpenReward’s leaderboard
| # | Entry | Score | Setting | Released | Run on | Link | Source |
|---|---|---|---|---|---|---|---|
| 1 | GPT-4V | 69.4% | Nov 2023 | Jul 2025 | arxiv.org | Official leaderboard | |
| 2 | GPT-4o Mar 2025 | 68% | Mar 2025 | Jul 2025 | arxiv.org | Official leaderboard | |
| 3 | GPT-4 Turbo Apr 2024 | 67.4% | Apr 2024 | Jul 2025 | arxiv.org | Official leaderboard | |
| 4 | ChipExpert* | 67.1% | Jul 2025 | arxiv.org | Official leaderboard | ||
| 5 | GPT-4 Jun 2023 | 63.7% | Jun 2023 | Jul 2025 | arxiv.org | Official leaderboard | |
| 6 | LlaMa3.2-Vision-Instruct-90B | 58.5% | Jul 2025 | arxiv.org | Official leaderboard | ||
| 7 | Reka Flash | 55.7% | Feb 2024 | Jul 2025 | arxiv.org | Official leaderboard | |
| 8 | GPT-3.5 Turbo Jan 2024 | 54.9% | Jan 2024 | Jul 2025 | arxiv.org | Official leaderboard | |
| 9 | QWen-2.5-Instruct-7B | 53% | Jul 2025 | arxiv.org | Official leaderboard | ||
| 10 | QWen-2-Instruct-72B | 50.2% | Jul 2025 | arxiv.org | Official leaderboard | ||
| 11 | QWen-2-Instruct-7B | 48.2% | Jul 2025 | arxiv.org | Official leaderboard | ||
| 12 | LlaMa3.2-Vision-Instruct-11B | 43.2% | Jul 2025 | arxiv.org | Official leaderboard | ||
| 13 | | 43% | May 2024 | Jul 2025 | arxiv.org | Official leaderboard | |
| 14 | | 41.7% | Jul 2024 | Jul 2025 | arxiv.org | Official leaderboard | |
| 15 | | 40.5% | Feb 2024 | Jul 2025 | arxiv.org | Official leaderboard | |
| 16 | DeepSeek-Math-7B-RL | 39.9% | Jul 2025 | arxiv.org | Official leaderboard | ||
| 17 | DeepSeek-V2-Lite-Chat | 39.7% | Jul 2025 | arxiv.org | Official leaderboard | ||
| 18 | | 38.8% | Jul 2024 | Jul 2025 | arxiv.org | Official leaderboard | |
| 19 | Reka Edge | 36.7% | Mar 2026 | Jul 2025 | arxiv.org | Official leaderboard | |
| 20 | DeepSeek-LLM-7B-Chat | 35.1% | Jul 2025 | arxiv.org | Official leaderboard | ||
| 21 | DeepSeek-MoE-16B-Chat | 34% | Jul 2025 | arxiv.org | Official leaderboard | ||
| 22 | | 32.5% | Dec 2023 | Jul 2025 | arxiv.org | Official leaderboard | |
| 23 | Qwen-VL-Chat | 32.2% | Aug 2023 | Jul 2025 | arxiv.org | Official leaderboard | |
| 24 | | 21.1% | Jan 2024 | Jul 2025 | arxiv.org | Official leaderboard | |
| 25 | InternLM-XComposer-VL-7B | 19.7% | Jul 2025 | arxiv.org | Official leaderboard | ||
| 26 | | 17.3% | Dec 2023 | Jul 2025 | arxiv.org | Official leaderboard | |
| 27 | | 14.3% | Feb 2023 | Jul 2025 | arxiv.org | Official leaderboard | |
| 28 | | 14% | Feb 2023 | Jul 2025 | arxiv.org | Official leaderboard | |
| 29 | QWen-2-Instruct-0.5B | 13.8% | Jul 2025 | arxiv.org | Official leaderboard | ||
| 30 | | 13.6% | Jun 2023 | Jul 2025 | arxiv.org | Official leaderboard | |
| 31 | | 11.5% | Apr 2024 | Jul 2025 | arxiv.org | Official leaderboard | |
| 32 | | 10.8% | May 2023 | Jul 2025 | arxiv.org | Official leaderboard | |
| 33 | | 10.7% | Jun 2023 | Jul 2025 | arxiv.org | Official leaderboard |
| # | Model | Score | Setting | Released | Run on | Link | Source |
|---|---|---|---|---|---|---|---|
| 1 | GPT-4V | 69.4% | Nov 2023 | Jul 2025 | arxiv.org | Official leaderboard | |
| 2 | GPT-4o Mar 2025 | 68% | Mar 2025 | Jul 2025 | arxiv.org | Official leaderboard | |
| 3 | GPT-4 Turbo Apr 2024 | 67.4% | Apr 2024 | Jul 2025 | arxiv.org | Official leaderboard | |
| 4 | GPT-4 Jun 2023 | 63.7% | Jun 2023 | Jul 2025 | arxiv.org | Official leaderboard | |
| 5 | Reka Flash | 55.7% | Feb 2024 | Jul 2025 | arxiv.org | Official leaderboard | |
| 6 | GPT-3.5 Turbo Jan 2024 | 54.9% | Jan 2024 | Jul 2025 | arxiv.org | Official leaderboard | |
| 7 | | 43% | May 2024 | Jul 2025 | arxiv.org | Official leaderboard | |
| 8 | | 41.7% | Jul 2024 | Jul 2025 | arxiv.org | Official leaderboard | |
| 9 | | 40.5% | Feb 2024 | Jul 2025 | arxiv.org | Official leaderboard | |
| 10 | | 38.8% | Jul 2024 | Jul 2025 | arxiv.org | Official leaderboard | |
| 11 | Reka Edge | 36.7% | Mar 2026 | Jul 2025 | arxiv.org | Official leaderboard | |
| 12 | | 32.5% | Dec 2023 | Jul 2025 | arxiv.org | Official leaderboard | |
| 13 | Qwen-VL-Chat | 32.2% | Aug 2023 | Jul 2025 | arxiv.org | Official leaderboard | |
| 14 | | 21.1% | Jan 2024 | Jul 2025 | arxiv.org | Official leaderboard | |
| 15 | | 17.3% | Dec 2023 | Jul 2025 | arxiv.org | Official leaderboard | |
| 16 | | 14.3% | Feb 2023 | Jul 2025 | arxiv.org | Official leaderboard | |
| 17 | | 14% | Feb 2023 | Jul 2025 | arxiv.org | Official leaderboard | |
| 18 | | 13.6% | Jun 2023 | Jul 2025 | arxiv.org | Official leaderboard | |
| 19 | | 11.5% | Apr 2024 | Jul 2025 | arxiv.org | Official leaderboard | |
| 20 | | 10.8% | May 2023 | Jul 2025 | arxiv.org | Official leaderboard | |
| 21 | | 10.7% | Jun 2023 | Jul 2025 | arxiv.org | Official leaderboard |
Rankings by part
We read 10 parts from OpenReward’s leaderboard, each ranking models on one part of the test, such as Cph, F and T. GPT-4V leads 5 of them, and 2 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.
Cph
21 models we track have a score here.
- 1 GPT-4o Mar 2025 ↑1 75.9%
- 2 GPT-4V ↓1 75.5%
- 3 GPT-4 Turbo Apr 2024 74%
- 4 GPT-4 Jun 2023 69.5%
- 5 Reka Flash 66.5%
- 6 GPT-3.5 Turbo Jan 2024 66.4%
- 7
MiniCPM-Llama3-V 2.5 53.5%
- 8
InternVL2-40B 51.8%
- 9
InternVL2-8B ↑1 47.1%
- 10 Qwen-VL-Chat ↑3 46.4%
F
21 models we track have a score here.
- 1 GPT-4V 79.8%
- 2 GPT-4 Turbo Apr 2024 ↑1 77.9%
- 3 GPT-4o Mar 2025 ↓1 76.2%
- 4 GPT-4 Jun 2023 74%
- 5 Reka Flash 68.3%
- 6 GPT-3.5 Turbo Jan 2024 66.7%
- 7
MiniCPM-Llama3-V 2.5 52.5%
- 8
InternVL2-40B 51.8%
- 9 Reka Edge ↑2 45.8%
- 10
InternVL2-8B 45.5%
T
21 models we track have a score here.
- 1 GPT-4 Turbo Apr 2024 ↑2 67.2%
- 2 GPT-4V ↓1 66.5%
- 3 GPT-4o Mar 2025 ↓1 65.5%
- 4 Reka Flash ↑1 63.4%
- 5 GPT-4 Jun 2023 ↓1 62%
- 6 GPT-3.5 Turbo Jan 2024 60.8%
- 7
InternVL2-40B ↑1 54.3%
- 8
InternVL2-8B ↑2 50.3%
- 9
DeepSeekMath-7B-Instruct 47.4%
- 10
MiniCPM-Llama3-V 2.5 ↓3 43.7%
K
21 models we track have a score here.
- 1 GPT-4 Turbo Apr 2024 ↑2 69.4%
- 2 GPT-4V ↓1 67.9%
- 3 GPT-4 Jun 2023 ↑1 66.7%
- 4 GPT-4o Mar 2025 ↓2 66.1%
- 5 Reka Flash 63%
- 6 GPT-3.5 Turbo Jan 2024 59.6%
- 7
InternVL2-40B ↑1 54.4%
- 8
InternVL2-8B ↑2 51.8%
- 9
DeepSeekMath-7B-Instruct 46.2%
- 10
MiniCPM-Llama3-V 2.5 ↓3 44.5%
S
21 models we track have a score here.
- 1 GPT-4o Mar 2025 ↑1 80.7%
- 2 GPT-4V ↓1 80.3%
- 3 GPT-4 Turbo Apr 2024 68.9%
- 4 GPT-4 Jun 2023 63.8%
- 5 GPT-3.5 Turbo Jan 2024 ↑1 58.6%
- 6 Reka Flash ↓1 54.7%
- 7
MiniCPM-Llama3-V 2.5 47.4%
- 8 Qwen-VL-Chat ↑5 41%
- 9
Yi-VL-34B ↑3 40.6%
- 10
DeepSeekMath-7B-Instruct ↓1 40.1%
G
21 models we track have a score here.
- 1 GPT-4V 69.9%
- 2 GPT-4o Mar 2025 69.4%
- 3 GPT-4 Turbo Apr 2024 67.9%
- 4 GPT-4 Jun 2023 64.7%
- 5 Reka Flash 61.8%
- 6
DeepSeekMath-7B-Instruct ↑3 53.3%
- 7 GPT-3.5 Turbo Jan 2024 ↓1 51.7%
- 8
InternVL2-8B ↑2 48.3%
- 9
InternVL2-40B ↓1 48.3%
- 10
MiniCPM-Llama3-V 2.5 ↓3 46.7%
R
21 models we track have a score here.
- 1 GPT-4V 67.3%
- 2 GPT-4 Turbo Apr 2024 ↑1 66.7%
- 3 GPT-4o Mar 2025 ↓1 66.3%
- 4 GPT-4 Jun 2023 61.3%
- 5 Reka Flash 61.3%
- 6 GPT-3.5 Turbo Jan 2024 58.4%
- 7
InternVL2-40B ↑1 45.7%
- 8
MiniCPM-Llama3-V 2.5 ↓1 44.5%
- 9 Reka Edge ↑2 41.9%
- 10
InternVL2-8B 39.5%
Cpt
21 models we track have a score here.
- 1 GPT-4V 59.2%
- 2 GPT-4o Mar 2025 56.9%
- 3 GPT-4 Turbo Apr 2024 54%
- 4
DeepSeekMath-7B-Instruct ↑5 46%
- 5 GPT-4 Jun 2023 ↓1 45.2%
- 6 GPT-3.5 Turbo Jan 2024 42.1%
- 7 Reka Flash ↓2 39.3%
- 8
InternVL2-40B 30.3%
- 9
InternVL2-8B ↑1 29%
- 10
MiniCPM-Llama3-V 2.5 ↓3 25.6%
B
21 models we track have a score here.
- 1 GPT-4 Turbo Apr 2024 ↑2 50.6%
- 2 GPT-4o Mar 2025 48.6%
- 3 GPT-4V ↓2 48.2%
- 4 GPT-4 Jun 2023 47.6%
- 5 GPT-3.5 Turbo Jan 2024 ↑1 37.2%
- 6 Reka Flash ↓1 33.5%
- 7
DeepSeekMath-7B-Instruct ↑2 32.3%
- 8
InternVL2-40B 25.7%
- 9
InternVL2-8B ↑1 24.8%
- 10
MiniCPM-Llama3-V 2.5 ↓3 24.3%
M
21 models we track have a score here.
- 1 GPT-4V 73.6%
- 2 GPT-4o Mar 2025 71.4%
- 3 GPT-4 Turbo Apr 2024 67.8%
- 4 GPT-4 Jun 2023 66.1%
- 5 GPT-3.5 Turbo Jan 2024 ↑1 46.7%
- 6 Reka Flash ↓1 44.8%
- 7
MiniCPM-Llama3-V 2.5 41.9%
- 8
Yi-VL-34B ↑4 31.8%
- 9
DeepSeekMath-7B-Instruct 30.8%
- 10 Reka Edge ↑1 29.3%
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: OpenReward’s leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: OpenReward’s 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
9 of 10 checks pass, 1 warning. Look first at the Newest score check. The newest came 8 months ago, in March 2026, for Reka Edge. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 3 pass 1 note
- Pass Independent runs Scores from the official leaderboard (33 entries).
- Pass Models scored 21 models we track, plus 12 other entries on its leaderboard.
- Warning Newest score The newest came 8 months ago, in March 2026, for Reka Edge.
- Note Frontier models None of the top 10 has a score. The strongest tested is GPT-4 Turbo (#172).
- Pass Headroom The best is 69.4%, under 85%, so it still separates models.
Leaderboard import 3 pass
- Pass Importer We read the board on OpenReward: 21 rows for models we track and 12 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.