Verilog Code Generation on RTLLM 2.0 (test) benchmark leaderboard · SOTA2 Research

Verilog Code Generation on RTLLM 2.0 (test)

74.45Pass@1VeriRL-CodeQwen2.5 + VERIMOA

Evaluation Results

Method	Links
VeriRL-CodeQwen2.5 + VERIMOA2025.10		74.45	78.63	82.02
Qwen2.5-Coder-32B + VERIMOA2025.10		65.49	71.42	74.11
VeriRL-CodeQwen2.52025.10		61.53	65.27	68.94
Qwen2.5-Coder-14B + VERIMOA2025.10		59.76	64.33	66.22
VeriRL-DeepSeek-Coder2025.10		58.64	63.92	66.26
Qwen2.5-Coder-7B + VERIMOA2025.10		54.43	62.52	66.24
HaVen-CodeQwen1.52025.10		51.04	57.38	60.89
OriGen-DeepSeek-Coder-7B-v1.52025.10		41.11	52.09	58.48
RTLCoder-DeepSeek-Coder2025.10		40.75	46.94	49.83
RTLCoder-Mistral2025.10		38.68	40.86	41.84