TuRTLe Code Completion (Verilator) — leaderboard
TuRTLe leaderboard variant for RTL code completion evaluated with Verilator across VerilogEval MC and VeriGen.
Metric: Aggregated Score (self-reported). Source: benchmarklist.com. Status: saturation imminent. 44 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GLM-5 | 82.23 |
| 2 | Gemma 4 31B | 80.51 |
| 3 | DeepSeek R1 0528 | 78.08 |
| 4 | DeepSeek V3.1 Terminus | 75.31 |
| 5 | GPT-OSS-120B | 74.91 |
| 6 | Qwen 3 235B A22B | 66.8 |
| 7 | GPT-OSS-20B | 65.92 |
| 8 | Llama 3.1 405B Instruct | 55.06 |
| 9 | Qwen 2.5 72B Instruct | 52.29 |
| 10 | Qwen 3 8B | 48.82 |
| 11 | Qwen 2.5 32B | 43.61 |
| 12 | QwQ-32B | 41.14 |
| 13 | starchat2-15B-v0.1 | 38.42 |
| 14 | DeepSeek R1 Distill Qwen 14B | 24.81 |
| 15 | Magistral Small | 23.47 |
Interactive version: theaggregate.ai/benchmark?slug=turtle-code-completion-verilator · How the rankings work · Data refreshed daily, snapshot 2026-07-22.