Benchmark Almanac

RTL-OPT

DeepSeek-Coder-V2-Lite-Instruct 72.5% leads since June 2024
Not in model cards

Models rewrite Verilog to improve a circuit while preserving its behavior. The authors compare synthesized power, timing, and area against human reference designs.

RTLBench tested five models on 40 released designs using equivalence and generic cell counts. The paper reports commercial synthesis results on 36 designs.

Kind
Electronics
Level
University
Scored
Checked automatically
Answers
Code
Tasks
All public
Measures
Efficiency
Language
English
In and out
Code → Code
Interaction
Single turn
Items from
Real work and use
Refresh
One fixed set
Run by
Anyone
Strongest model tested
Qwen3.6 27B #88 on the capability index
Last model added
April 2026
Scores refreshed
November 20th, 2026
5
Models scored on its own leaderboard

DeepSeek-Coder-V2-Lite-Instruct leads with 72.5% among models released by April 2026

Best score per model from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best.

The newest model with a score here came out in April 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:

  1. 1 DeepSeek-Coder-V2-Lite-Instruct 72.5%
  2. 2 Qwen3.6 27B 62.5%
  3. 3 Qwen2.5-Coder-32B-Instruct 47.5%
  4. 4 Qwen3-Coder-30B-A3B-Instruct 47.5%
  5. 5 Qwen 3.6 35B-A3B 47.5%

Source: Official RTL-OPT leaderboard

#ModelScoreSettingDateSource
1 DeepSeek-Coder-V2-Lite-Instruct 72.5% Jun 2024 Official leaderboard
2 Qwen3.6 27B 62.5% Apr 2026 Official leaderboard
3 Qwen2.5-Coder-32B-Instruct 47.5% Nov 2024 Official leaderboard
4 Qwen3-Coder-30B-A3B-Instruct 47.5% Jul 2025 Official leaderboard
5 Qwen 3.6 35B-A3B 47.5% Apr 2026 Official leaderboard

Number one over time

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

Other labs
Own leaderboard Not tested since Apr 2026

Source: Official RTL-OPT 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
0% 20% 40% 60% 80% 100% 2024 2025 2026 Near the ceiling (85%) DeepSeek-Coder-V2-Lite-Instruct 0% 20% 40% 60% 80% 100% 2024 2025 2026 Near the ceiling (85%) DeepSeek-Coder-V2-Lite-Instruct

Source: Official RTL-OPT 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 7 months ago, in April 2026, for Qwen3.6 27B. 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 (5 entries).
  • Pass Models scored 5 models we track.
  • Warning Newest score The newest came 7 months ago, in April 2026, for Qwen3.6 27B.
  • Note Frontier models None of the top 10 has a score. The strongest tested is Qwen3.6 27B (#88).
  • Pass Headroom The best is 72.5%, under 85%, so it still separates models.
Leaderboard import 3 pass
  • Pass Importer We read the board on github.com: 5 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.

How we check