# PennyCoder: Efficient Domain-Specific LLMs for PennyLane-Based Quantum Code Generation

arXiv: https://arxiv.org/abs/2507.19562 · HTML: https://mlsanity.com/paper/2507.19562/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 15 of its 27 references that are in the map.
- First posted on arXiv 25 Jul 2025.

## Builds on (most cited first, 15 of 15)

- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Hierarchical Text-Conditional Image Generation with CLIP Latents](https://mlsanity.com/paper/2204.06125/) (2022)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Reflexion: Language Agents with Verbal Reinforcement Learning](https://mlsanity.com/paper/2303.11366/) (2023)
- [Program Synthesis with Large Language Models](https://mlsanity.com/paper/2108.07732/) (2021)
- [StarCoder: may the source be with you!](https://mlsanity.com/paper/2305.06161/) (2023)
- [Quantum Long Short-Term Memory](https://mlsanity.com/paper/2009.01783/) (2020)
- [Qiskit HumanEval: An Evaluation Benchmark For Quantum Code Generative Models](https://mlsanity.com/paper/2406.14712/) (2024)
- [A Survey on Quantum Machine Learning: Current Trends, Challenges, Opportunities, and the Road Ahead](https://mlsanity.com/paper/2310.10315/) (2023)
- [Qiskit Code Assistant: Training LLMs for generating Quantum Computing Code](https://mlsanity.com/paper/2405.19495/) (2024)
- [The power and limitations of learning quantum dynamics incoherently](https://mlsanity.com/paper/2303.12834/) (2023)
- [A PennyLane-Centric Dataset to Enhance LLM-based Quantum Code Generation using RAG](https://mlsanity.com/paper/2503.02497/) (2025)
- [QHackBench: Benchmarking Large Language Models for Quantum Code Generation Using PennyLane Hackathon Challenges](https://mlsanity.com/paper/2506.20008/) (2025)
- [An Introduction to Quantum Reinforcement Learning (QRL)](https://mlsanity.com/paper/2409.05846/) (2024)

## Built on by (most cited first, 6 of 6)

- [Quantum Verifiable Rewards for Post-Training Qiskit Code Assistant](https://mlsanity.com/paper/2508.20907/) (2025)
- [QuanBench+: A Unified Multi-Framework Benchmark for LLM-Based Quantum Code Generation](https://mlsanity.com/paper/2604.08570/) (2026)
- [Qiskit QuantumKatas: Adapting Microsoft's Quantum Computing exercises for LLM evaluation](https://mlsanity.com/paper/2605.27210/) (2026)
- [QBugLM: An Agentic Benchmarking Framework for LLM-based Quantum Software Debugging](https://mlsanity.com/paper/2606.07314/) (2026)
- [PennySynth: RAG-Driven Data Synthesis for Automated Quantum Code Generation](https://mlsanity.com/paper/2605.25572/) (2026)
- [Fine-Tuning Large Language Models for Quantum Reasoning](https://mlsanity.com/paper/2606.21974/) (2026)

## Abstract (arXiv)

The growing demand for robust quantum programming frameworks has unveiled a critical limitation: current large language model (LLM) based quantum code assistants heavily rely on remote APIs, introducing challenges related to privacy, latency, and excessive usage costs. Addressing this gap, we propose PennyCoder, a novel lightweight framework for quantum code generation, explicitly designed for local and embedded deployment to enable on-device quantum programming assistance without external API dependence. PennyCoder leverages a fine-tuned version of the LLaMA 3.1-8B model, adapted through parameter-efficient Low-Rank Adaptation (LoRA) techniques combined with domain-specific instruction tuning optimized for the specialized syntax and computational logic of quantum programming in PennyLane, including tasks in quantum machine learning and quantum reinforcement learning. Unlike prior work focused on cloud-based quantum code generation, our approach emphasizes device-native operability while maintaining high model efficacy. We rigorously evaluated PennyCoder over a comprehensive quantum programming dataset, achieving 44.3% accuracy with our fine-tuned model (compared to 33.7% for the base LLaMA 3.1-8B and 40.1% for the RAG-augmented baseline), demonstrating a significant improvement in functional correctness.