TuRTLe Spec-to-RTL (Verilator) — leaderboard
TuRTLe leaderboard variant for spec-to-RTL generation evaluated with Verilator across VerilogEval S2R and RTLLM.
Metric: Aggregated Score (self-reported). Source: benchmarklist.com. Status: saturation imminent. 44 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemma 4 31B | 79.83 |
| 2 | GLM-5 | 78.17 |
| 3 | DeepSeek R1 0528 | 75.83 |
| 4 | DeepSeek V3.1 Terminus | 71.35 |
| 5 | GPT-OSS-120B | 70.18 |
| 6 | Qwen 3 235B A22B | 69.17 |
| 7 | QwQ-32B | 63.75 |
| 8 | GPT-OSS-20B | 63.2 |
| 9 | Qwen 2.5 32B | 53.2 |
| 10 | Llama 3.1 405B Instruct | 52.08 |
| 11 | Qwen 2.5 72B Instruct | 51.72 |
| 12 | Qwen 3 8B | 46.23 |
| 13 | Magistral Small | 41.02 |
| 14 | starchat2-15B-v0.1 | 40.2 |
| 15 | DeepSeek R1 Distill Qwen 14B | 22.93 |
Interactive version: theaggregate.ai/benchmark?slug=turtle-spec-to-rtl-verilator · How the rankings work · Data refreshed daily, snapshot 2026-07-22.