List your company
ResearchFormerly WizWand

RTL Generation on RTLLM VerilogEval CVDP Combined v2

94Mean Z-score Pass@10GPT-OSS-120B
-211.76-132.38-5326.38Jun 11, 2026
Updated 4mo ago

Evaluation results

MethodLinks
GPT-OSS-120BRole=Teacher, SFT=null, RL=null2026.0694
STGRole=Fine-tuned, Base Model=Qwen3-8B, SFT=true, RL=false2026.0677
STGRole=Fine-tuned, Base Model=Qwen3-4B-Thinking, SFT=true, RL=false2026.0668
STGRole=Fine-tuned, Base Model=Qwen2.5-C-7B-Instruct, SFT=true, RL=false2026.0656
CodeV-R1Role=Fine-tuned, Base Model=Qwen2.5-C-7B-Instruct, SFT=true, RL=true2026.0654
SiliconMind-V1Role=Fine-tuned, Base Model=Qwen3-8B, SFT=true, RL=false2026.0646
SiliconMind-V1Role=Fine-tuned, Base Model=Qwen3-4B-Thinking, SFT=true, RL=false2026.0637
SiliconMind-V1Role=Fine-tuned, Base Model=Qwen2.5-C-7B-Instruct, SFT=true, RL=false2026.0630
VeriRL (reproduced)Role=Fine-tuned, Base Model=Qwen2.5-C-7B-Instruct, SFT=true, RL=true2026.06-29
Qwen3-8BRole=Base, SFT=null, RL=null2026.06-80
Qwen2.5-C-7B-InstructRole=Base, SFT=null, RL=null2026.06-151
Qwen3-4B-ThinkingRole=Base, SFT=null, RL=null2026.06-200