Benchmark Almanac

AnalogCoder

Claude-3.5 91.7% tops its leaderboard Llama 3.1-405B 83.3% leads the models we track
Not in model cards

Models write Python programs that build analog circuits. A simulator checks whether each circuit performs its requested function. The authors report first-attempt success and how many tasks each model solves.

The authors publish results on 24 circuits. Their repository has 213 stars. The ranking includes model variants with different prompts and tools.

Made by
Yao Lai and colleagues
Released
May 2024
Kind
Electronics
Level
University
Scored
Checked automatically
Answers
Code
Tasks
All public
Measures
Capability
Language
English
In and out
Text → Code
Interaction
Agentic
Environment
A simulator
Items from
Written by experts
Refresh
One fixed set
Run by
Anyone
In views
Agents
Strongest model tested
Gemini 1.5 Pro Sep 2024 #157 on the capability index
Last model added
March 2025
Scores refreshed
November 20th, 2026
40
Entries on its own leaderboard, 27 models we track

Claude-3.5 tops the leaderboard with 91.7% Llama 3.1-405B leads with 83.3% among models released by March 2025

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 we track is Llama 3.1-405B, at 83.3%. Agents built around a model, such as AnalogCoder + GPT-4o, 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 March 2025, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 Claude-3.5 91.7%
  2. 2 Llama 3.1-405B 83.3%
  3. 3 AnalogCoder + GPT-4o Mar 2025 83.3%
  4. 4 DeepSeek-Coder-V2 236B 79.2%
  5. 5 Gemini 1.5 Pro Sep 2024 70.8%
  6. 6 Mistral Large 2 Nov 2024 70.8%
  7. 7 Claude 3.5 Sonnet w/o tool 62.5%
  8. 8 GPT-4o Mar 2025, w/o tool 62.5%
  9. 9 GPT-4 Jun 2023, w/o tool 58.3%
  10. 10 Llama 3.1-70B 58.3%
  11. 11 DeepSeek-V2 (MoE-236B, May 2024) 54.2%
  12. 12 Qwen2-72B 50%
  13. 13 GPT-4o mini 50%
  14. 14 Gemini 1.0 Pro 45.8%
  15. 15 Llama 3-70B 45.8%
  16. 16 Gemini 1.5 Flash Sep 2024 45.8%
  17. 17 GPT-3.5 41.7%
  18. 18 GLM-4 33.3%
  19. 19 Codestral 33.3%
  20. 20 CodeGemma-7B 29.2%
  21. 21 CodeLlama-70B 29.2%
  22. 22 WizardCoder-33B 29.2%
  23. 23 CodeGeeX4-9B 25%
  24. 24 Magicoder-7B 20.8%
  25. 25 StarCoder2-15B-Instruct 20.8%
  26. 26 Mixtral 8x7B 20.8%
  27. 27 CodeLlama-34B 16.7%
  28. 28 CodeLlama-7B 16.7%
  29. 29 DeepSeek Coder 33B 16.7%
  30. 30 CodeQwen1.5-7B 16.7%
  31. 31 Llama 3.1-8B 16.7%
  32. 32 Llama 2-70B 12.5%
  33. 33 CodeLlama-13B 8.33%
  34. 34 Mistral 7B v0.1 8.33%
  35. 35 Qwen1.5-110B 8.33%
  36. 36 SemiKong-8B 4.17%
  37. 37 Llama 3-8B 4.17%
  38. 38 phi-3-medium 14B 4.17%
  39. 39 Llama 2-13B 0%
  40. 40 Llama 2-7B 0%

Source: Official AnalogCoder leaderboard

#EntryScoreSettingDateSource
1 Claude-3.5 91.7% AnalogCoder Official leaderboard
2 Llama 3.1-405B 83.3% Jul 2024 Official leaderboard
3 AnalogCoder + GPT-4o Mar 2025 83.3% Mar 2025 Official leaderboard
4 DeepSeek-Coder-V2 236B 79.2% Jun 2024 Official leaderboard
5 Gemini 1.5 Pro Sep 2024 70.8% Sep 2024 Official leaderboard
6 Mistral Large 2 Nov 2024 70.8% Nov 2024 Official leaderboard
7 Claude 3.5 Sonnet 62.5% w/o tool Jun 2024 Official leaderboard
8 GPT-4o Mar 2025 62.5% w/o tool Mar 2025 Official leaderboard
9 GPT-4 Jun 2023 58.3% w/o tool Jun 2023 Official leaderboard
10 Llama 3.1-70B 58.3% Jul 2024 Official leaderboard
11 DeepSeek-V2 (MoE-236B, May 2024) 54.2% May 2024 Official leaderboard
12 Qwen2-72B 50% Jun 2024 Official leaderboard
13 GPT-4o mini 50% Jul 2024 Official leaderboard
14 Gemini 1.0 Pro 45.8% Dec 2023 Official leaderboard
15 Llama 3-70B 45.8% Apr 2024 Official leaderboard
16 Gemini 1.5 Flash Sep 2024 45.8% Aug 2024 Official leaderboard
17 GPT-3.5 41.7% Official leaderboard
18 GLM-4 33.3% Jan 2024 Official leaderboard
19 Codestral 33.3% Jan 2025 Official leaderboard
20 CodeGemma-7B 29.2% Official leaderboard
21 CodeLlama-70B 29.2% Official leaderboard
22 WizardCoder-33B 29.2% Official leaderboard
23 CodeGeeX4-9B 25% Official leaderboard
24 Magicoder-7B 20.8% Official leaderboard
25 StarCoder2-15B-Instruct 20.8% Official leaderboard
26 Mixtral 8x7B 20.8% Dec 2023 Official leaderboard
27 CodeLlama-34B 16.7% Official leaderboard
28 CodeLlama-7B 16.7% Official leaderboard
29 DeepSeek Coder 33B 16.7% Nov 2023 Official leaderboard
30 CodeQwen1.5-7B 16.7% Apr 2024 Official leaderboard
31 Llama 3.1-8B 16.7% Jul 2024 Official leaderboard
32 Llama 2-70B 12.5% Jul 2023 Official leaderboard
33 CodeLlama-13B 8.33% Official leaderboard
34 Mistral 7B v0.1 8.33% Sep 2023 Official leaderboard
35 Qwen1.5-110B 8.33% Apr 2024 Official leaderboard
36 SemiKong-8B 4.17% Official leaderboard
37 Llama 3-8B 4.17% Apr 2024 Official leaderboard
38 phi-3-medium 14B 4.17% Apr 2024 Official leaderboard
39 Llama 2-13B 0% Jul 2023 Official leaderboard
40 Llama 2-7B 0% Jul 2023 Official leaderboard
#ModelScoreSettingDateSource
1 Llama 3.1-405B 83.3% Jul 2024 Official leaderboard
2 DeepSeek-Coder-V2 236B 79.2% Jun 2024 Official leaderboard
3 Gemini 1.5 Pro Sep 2024 70.8% Sep 2024 Official leaderboard
4 Mistral Large 2 Nov 2024 70.8% Nov 2024 Official leaderboard
5 Claude 3.5 Sonnet 62.5% w/o tool Jun 2024 Official leaderboard
6 GPT-4o Mar 2025 62.5% w/o tool Mar 2025 Official leaderboard
7 GPT-4 Jun 2023 58.3% w/o tool Jun 2023 Official leaderboard
8 Llama 3.1-70B 58.3% Jul 2024 Official leaderboard
9 DeepSeek-V2 (MoE-236B, May 2024) 54.2% May 2024 Official leaderboard
10 Qwen2-72B 50% Jun 2024 Official leaderboard
11 GPT-4o mini 50% Jul 2024 Official leaderboard
12 Gemini 1.0 Pro 45.8% Dec 2023 Official leaderboard
13 Llama 3-70B 45.8% Apr 2024 Official leaderboard
14 Gemini 1.5 Flash Sep 2024 45.8% Aug 2024 Official leaderboard
15 GLM-4 33.3% Jan 2024 Official leaderboard
16 Codestral 33.3% Jan 2025 Official leaderboard
17 Mixtral 8x7B 20.8% Dec 2023 Official leaderboard
18 DeepSeek Coder 33B 16.7% Nov 2023 Official leaderboard
19 CodeQwen1.5-7B 16.7% Apr 2024 Official leaderboard
20 Llama 3.1-8B 16.7% Jul 2024 Official leaderboard
21 Llama 2-70B 12.5% Jul 2023 Official leaderboard
22 Mistral 7B v0.1 8.33% Sep 2023 Official leaderboard
23 Qwen1.5-110B 8.33% Apr 2024 Official leaderboard
24 Llama 3-8B 4.17% Apr 2024 Official leaderboard
25 phi-3-medium 14B 4.17% Apr 2024 Official leaderboard
26 Llama 2-13B 0% Jul 2023 Official leaderboard
27 Llama 2-7B 0% Jul 2023 Official leaderboard

3 models have held number one on the official leaderboard since its release in May 2024

Lab claims count from the day the card came out, independent runs from the model’s release.

OpenAI DeepSeek Meta
Own leaderboard Not tested since Mar 2025

Source: Official AnalogCoder leaderboard

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

Plot with other benchmarks
On the benchmark’s own leaderboard Best so far
0% 20% 40% 60% 80% 100% 2023 2024 2025 Near the ceiling (85%) Llama 3.1-405B DeepSeek-Coder-V2 236B GPT-4 0% 20% 40% 60% 80% 100% 2023 2024 2025 Near the ceiling (85%) Llama 3.1-405B GPT-4

Source: Official AnalogCoder 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

All 112 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 20 months ago, in March 2025, for GPT-4o. 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 (40 entries).
  • Pass Models scored 27 models we track, plus 13 other entries on its leaderboard.
  • Warning Newest score The newest came 20 months ago, in March 2025, for GPT-4o.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Gemini 1.5 Pro (#157).
  • Pass Headroom The best is 83.3%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on github.com: 28 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.

How we check