CktEvo: leaderboard
Metric: Reduction (%) of the geometric-mean area-delay product over CktEvo's 11 Verilog repositories relative to the golden designs, after the model drives the authors' island-based evolutionary loop (Yosys synthesis feedback, graph-based bottleneck analysis, formal equivalence checking of every accepted revision); higher is better. Source: arxiv.org. Saturation forecast: Around 2028. 3 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4o | 9.23 | #333 |
No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #8 Claude Fable 5.1.
Interactive version: theaggregate.ai/benchmark?slug=cktevo · How It Works · Data refreshed daily, snapshot 2026-10-11.