CodeEvolve: an open source evolutionary coding agent for algorithmic discovery and optimization
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911727778529280 |
|---|---|
| author | Assumpção, Henrique Ferreira, Diego Campos, Leandro Murai, Fabricio |
| author_facet | Assumpção, Henrique Ferreira, Diego Campos, Leandro Murai, Fabricio |
| contents | We introduce CodeEvolve, an open-source framework that couples large language models with island-based evolutionary search for end-to-end algorithmic discovery. CodeEvolve integrates inspiration-based crossover, meta-prompting, and depth-based refinement on top of a CVT-MAP-Elites archive and a weighted LLM ensemble to generate optimized solutions for complex problems. On the AlphaEvolve benchmark suite, CodeEvolve matches or surpasses the reported AlphaEvolve results on 5 of 9 problems and, under matched conditions, outperforms the open-source frameworks OpenEvolve and ShinkaEvolve on 6 of 9. With the open-weight Qwen3-Coder-30B backbone, it surpasses the reported AlphaEvolve score on both CirclePackingSquare instances at roughly an order of magnitude lower cost than a frontier closed-source ensemble, and remains competitive with EoH on heuristic-design tasks without retuning. Ablations show that the interaction between CodeEvolve's components, rather than any single operator, drives these results. We release the framework, experimental data, and practical hyperparameter guidelines at https://github.com/inter-co/science-codeevolve. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_14150 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | CodeEvolve: an open source evolutionary coding agent for algorithmic discovery and optimization Assumpção, Henrique Ferreira, Diego Campos, Leandro Murai, Fabricio Artificial Intelligence Machine Learning Neural and Evolutionary Computing I.2.7; I.2.2 We introduce CodeEvolve, an open-source framework that couples large language models with island-based evolutionary search for end-to-end algorithmic discovery. CodeEvolve integrates inspiration-based crossover, meta-prompting, and depth-based refinement on top of a CVT-MAP-Elites archive and a weighted LLM ensemble to generate optimized solutions for complex problems. On the AlphaEvolve benchmark suite, CodeEvolve matches or surpasses the reported AlphaEvolve results on 5 of 9 problems and, under matched conditions, outperforms the open-source frameworks OpenEvolve and ShinkaEvolve on 6 of 9. With the open-weight Qwen3-Coder-30B backbone, it surpasses the reported AlphaEvolve score on both CirclePackingSquare instances at roughly an order of magnitude lower cost than a frontier closed-source ensemble, and remains competitive with EoH on heuristic-design tasks without retuning. Ablations show that the interaction between CodeEvolve's components, rather than any single operator, drives these results. We release the framework, experimental data, and practical hyperparameter guidelines at https://github.com/inter-co/science-codeevolve. |
| title | CodeEvolve: an open source evolutionary coding agent for algorithmic discovery and optimization |
| topic | Artificial Intelligence Machine Learning Neural and Evolutionary Computing I.2.7; I.2.2 |
| url | https://arxiv.org/abs/2510.14150 |