RealBench: Benchmarking Verilog Generation Models with Real-World IP Designs
By: Pengwei Jin , Di Huang , Chongxiao Li and more
Potential Business Impact:
Helps computers write complex computer parts.
The automatic generation of Verilog code using Large Language Models (LLMs) has garnered significant interest in hardware design automation. However, existing benchmarks for evaluating LLMs in Verilog generation fall short in replicating real-world design workflows due to their designs' simplicity, inadequate design specifications, and less rigorous verification environments. To address these limitations, we present RealBench, the first benchmark aiming at real-world IP-level Verilog generation tasks. RealBench features complex, structured, real-world open-source IP designs, multi-modal and formatted design specifications, and rigorous verification environments, including 100% line coverage testbenches and a formal checker. It supports both module-level and system-level tasks, enabling comprehensive assessments of LLM capabilities. Evaluations on various LLMs and agents reveal that even one of the best-performing LLMs, o1-preview, achieves only a 13.3% pass@1 on module-level tasks and 0% on system-level tasks, highlighting the need for stronger Verilog generation models in the future. The benchmark is open-sourced at https://github.com/IPRC-DIP/RealBench.
Similar Papers
ResBench: Benchmarking LLM-Generated FPGA Designs with Resource Awareness
Hardware Architecture
Helps computers design better chips faster.
Insights from Verification: Training a Verilog Generation LLM with Reinforcement Learning with Testbench Feedback
Hardware Architecture
Makes computer chips work right from descriptions.
ArchXBench: A Complex Digital Systems Benchmark Suite for LLM Driven RTL Synthesis
Hardware Architecture
AI designs complex computer chips automatically.