# Improving HPC Code Generation Capability of LLMs via Online Reinforcement Learning with Real-Machine Benchmark Rewards

arXiv: https://arxiv.org/abs/2602.12049 · HTML: https://mlsanity.com/paper/2602.12049/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 31 references that are in the map.
- First posted on arXiv 12 Feb 2026.

## Builds on (most cited first, 24 of 24)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [Code Llama: Open Foundation Models for Code](https://mlsanity.com/paper/2308.12950/) (2023)
- [Qwen2.5-Coder Technical Report](https://mlsanity.com/paper/2409.12186/) (2024)
- [KernelBench: Can LLMs Write Efficient GPU Kernels?](https://mlsanity.com/paper/2502.10517/) (2025)
- [RLEF: Grounding Code LLMs in Execution Feedback with Reinforcement Learning](https://mlsanity.com/paper/2410.02089/) (2024)
- [RLTF: Reinforcement Learning from Unit Test Feedback](https://mlsanity.com/paper/2307.04349/) (2023)
- [StepCoder: Improve Code Generation with Reinforcement Learning from Compiler Feedback](https://mlsanity.com/paper/2402.01391/) (2024)
- [Can Large Language Models Write Parallel Code?](https://mlsanity.com/paper/2401.12554/) (2024)
- [Mercury: A Code Efficiency Benchmark for Code Large Language Models](https://mlsanity.com/paper/2402.07844/) (2024)
- [Performance-Aligned LLMs for Generating Fast Code](https://mlsanity.com/paper/2404.18864/) (2024)
- [CUDA-L2: Surpassing cuBLAS Performance for Matrix Multiplication through Reinforcement Learning](https://mlsanity.com/paper/2512.02551/) (2025)
- [AscendKernelGen: A Systematic Study of LLM-Based Kernel Generation for Neural Processing Units](https://mlsanity.com/paper/2601.07160/) (2026)
- [MARCO: Multi-Agent Code Optimization with Real-Time Knowledge Integration for High-Performance Computing](https://mlsanity.com/paper/2505.03906/) (2025)
- [Agentic Auto-Scheduling: An Experimental Study of LLM-Guided Loop Optimization](https://mlsanity.com/paper/2511.00592/) (2025)
- [MaxCode: A Max-Reward Reinforcement Learning Framework for Automated Code Optimization](https://mlsanity.com/paper/2601.05475/) (2026)
- [LLM-HPC++: Evaluating LLM-Generated Modern C++ and MPI+OpenMP Codes for Scalable Mandelbrot Set Computation](https://mlsanity.com/paper/2512.17023/) (2025)
- [A Two-Stage GPU Kernel Tuner Combining Semantic Refactoring and Search-Based Optimization](https://mlsanity.com/paper/2601.12698/) (2026)
- [Pearl: Automatic Code Optimization Using Deep Reinforcement Learning](https://mlsanity.com/paper/2506.01880/) (2025)
- [VibeCodeHPC: An Agent-Based Iterative Prompting Auto-Tuner for HPC Code Generation Using LLMs](https://mlsanity.com/paper/2510.00031/) (2025)
- [ContextEvolve: Multi-Agent Context Compression for Systems Code Optimization](https://mlsanity.com/paper/2602.02597/) (2026)

## Built on by (most cited first, 1 of 1)

- [Reinforcement Learning for Code Optimization](https://mlsanity.com/paper/2607.25970/) (2026)

## Abstract (arXiv)

Large language models (LLMs) have demonstrated strong code generation capabilities, yet the runtime performance of generated code is not guaranteed, and there have been few attempts to train LLMs using runtime performance as a reward in the HPC domain. We propose an online reinforcement learning approach that executes LLM-generated code on a supercomputer and directly feeds back the measured runtime performance (GFLOPS) as a reward. We further introduce a Staged Quality-Diversity (SQD) algorithm that progressively varies the permitted optimization techniques on a per-problem basis, enabling the model to learn code optimization from diverse perspectives. We build a distributed system connecting a GPU training cluster with a CPU benchmarking cluster, and train Qwen2.5 Coder 14B on a double-precision matrix multiplication task using Group Relative Policy Optimization (GRPO). Through two experiments, we show that reinforcement learning combining runtime performance feedback with staged optimization can improve the HPC code generation capability of LLMs.