zhuyaoyu/CodeV-R1-RL-Qwen-7B

CodeV-R1-Qwen-7B is a 7.6 billion parameter language model developed by Yaoyu Zhu et al., fine-tuned for Verilog generation using a novel reinforcement learning with verifiable reward (RLVR) framework. Built upon the Qwen-2.5 series, it excels at translating natural language specifications into Verilog and Verilog code completion, demonstrating superior efficiency and performance on hardware description language tasks compared to general-purpose and other coding LLMs. This model is specifically optimized for electronic design automation (EDA) applications, addressing challenges in verification, data scarcity, and computational cost.