Note: This repository contains the source code for our anonymous submission to an ASP-DAC 2026. The contents have been cleaned to remove any personally identifiable information.

Large Language Models (LLMs) are used for Register-Transfer Level (RTL) code generation, but they face two main challenges: functional correctness and Power, Performance, and Area (PPA) optimization. Iterative, feedback-based methods partially address these, but they are limited to local search, hindering the discovery of a global optimum. This paper introduces REvolution, a framework that combines Evolutionary Computation (EC) with LLMs for automatic RTL generation and optimization. REvolution evolves a population of candidates in parallel, each defined by a design strategy (Thought), RTL implementation (Code), and evaluation feedback. The framework includes a dual-population algorithm that divides candidates into Fail and Success groups for bug fixing and PPA optimization, respectively. An adaptive mechanism further improves search efficiency by dynamically adjusting the selection probability according to the success rates. Experiments on the VerilogEval and RTLLM benchmarks show that REvolution increased the initial pass rate of various LLMs by up to 24.0 percentage points. The DeepSeekV3 model achieved a final pass rate of 95.5%, comparable to state-of-the-art results, without the need for separate training or domain-specific tools. Additionally, the generated RTL designs showed significant PPA improvements over reference designs. This work introduces a new RTL design paradigm by combining LLMs' generative capabilities with EC's broad search power, overcoming the local-search limitations of previous methods.

The repository is organized as follows:

• ./script/main.py: The main script that contains the full implementation of the REvolution framework.
• ./exp/: This directory contains all experimental results. The _clean suffix in the directory names indicates that potentially identifiable information has been removed for anonymous review.
  • ./exp/llama3_clean_results/: Results from the REvolution run using Llama-3.3-70B.
  • ./exp/llama3_baseline_clean_results/: Baseline results for Llama-3.3-70B (pass@200) without the evolutionary framework.
  • ./exp/gpt-4.1-mini_clean_results/: Results from the REvolution run using GPT-4.1-mini.
  • ./exp/deepseek_clean_results/: Results from the REvolution run using DeepSeek-V3-0324.

Follow these steps to set up the required environment.

First, ensure you have the following tools installed and available in your system's PATH. The versions used in our experiments are specified below:

• Yosys: 0.54+29
• Icarus Verilog: v12_0
• OpenROAD: v2.0-22560-gb571c4b471

The Python environment is managed using uv.

1. Install uv: If you don't have uv, install it by following the official instructions.
2. Create Virtual Environment: Run the following command in the repository root to create and sync the virtual environment using the provided lock file.

   uv sync

3. Activate Environment: Activate the newly created environment.

   source .venv/bin/activate

The framework supports multiple LLM backends.

• External APIs (OpenAI, DeepSeek, OpenRouter): Set the appropriate environment variable with your API key.

  # For OpenAI
  export OPENAI_API_KEY="your-key-here"
  
  # For DeepSeek
  export DEEPSEEK_API_KEY="your-key-here"
  
  # For OpenRouter
  export OPENROUTER_API_KEY="your-key-here"

• Local vLLM Server: If you're using a local vLLM instance, ensure it's running and accessible at http://localhost:8000/v1. No API key is needed for this setup.

The following commands can be used to reproduce the experiments presented in the paper. Adjust the --num_workers argument based on your available computing resources.

This command runs the baseline experiment without the evolutionary framework to calculate pass@200.

python3 script/main.py \
    --strategy_selection ucb \
    --api_backend openrouter \
    --model_name meta-llama/llama-3.3-70b-instruct \
    --num_workers 32 \
    --num_generations 0 \
    --benchmarks VerilogEval-Spec-to-RTL RTLLM \
    --max_tokens 4096 \
    --temperature 1.0 \
    --top_p 0.95 \
    --population_size 200

These commands execute the REvolution framework for different models.

Llama-3.3-70B-Instruct

python3 script/main.py \
    --strategy_selection ucb \
    --api_backend openrouter \
    --model_name meta-llama/llama-3.3-70b-instruct \
    --num_workers 48 \
    --num_generations 20 \
    --benchmarks VerilogEval-Spec-to-RTL \
    --max_tokens 4096 \
    --temperature 1.0 \
    --top_p 0.95 \
    --population_size 10

DeepSeek-V3-0324

python3 script/main.py \
    --strategy_selection ucb \
    --api_backend deepseek \
    --model_name deepseek-chat \
    --num_workers 20 \
    --num_generations 20 \
    --benchmarks RTLLM VerilogEval-Spec-to-RTL \
    --max_tokens 4096 \
    --temperature 1.0 \
    --top_p 0.95 \
    --population_size 10

GPT-4.1-mini

python3 script/main.py \
    --strategy_selection ucb \
    --model_name gpt-4.1-mini \
    --num_workers 32 \
    --num_generations 20 \
    --benchmarks RTLLM VerilogEval-Spec-to-RTL \
    --max_tokens 4096 \
    --temperature 1.0 \
    --top_p 0.95 \
    --population_size 10

After running the experiments, you can generate reports to view the results.

To generate a detailed markdown report for a specific experimental run, use the evolutionary_report_generator.py script. This report includes PPA metrics but does not apply the gate-level cutoff used in the paper.

python script/evolutionary_report_generator.py \
    --experiment_path ./exp/deepseek_clean_results \
    --save_markdown

This will save a report named (benchmark_name)_evolutionary_report.md inside the specified experiment directory.

To generate the final, PPA-filtered results as reported in our paper, run the following shell script. It applies a gate count (50) cutoff to filter the results.

./script/generate_cutoff_compile_result_variants.sh --gate 50

This will create a compile_results_gate_cutoff_50.md file in the base directory, containing the table of results presented in the paper.

To recreate the PPA scatterplot for the VerilogEval-Spec-to-RTL/Prob033_ece241_2014_q1c problem, run the following script:

python3 script/plot_problem_pareto.py

This script will generate the plots and save them in a newly created directory named VerilogEval_Prob033_ece241_2014_q1c_plots.