mlsanityData to 14 Oct 2026

CodeEvolve: an open source evolutionary coding agent for algorithmic discovery and optimization

arXiv 2510.14150 ↗First posted 15 Oct 2025cs.AI cs.LG cs.NEcited by 44

On the map: LLM agents and tool use → Software engineering coding agents

Abstract (arXiv)

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.

Key facts. Cited by 44 papers in the map (data through 14 Oct 2026). Placed in Software engineering coding agents, one of 16 directions in LLM agents and tool use. Builds on 17 of its 58 references that are in the map. First posted on arXiv 15 Oct 2025.

Builds on

17 of 58 references.

Closest ancestors · 1

Same direction, newest first

  1. 2025AlphaEvolve: A coding agent for scientific and algorithmic discovery16 Jun

Other references · 16

  1. 2021Evaluating Large Language Models Trained on Codecited by 8,320
  2. 2025Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilitiescited by 3,796
  3. 2023Self-Refine: Iterative Refinement with Self-Feedbackcited by 3,442
  4. 2015Illuminating search spaces by mapping elitescited by 425
  5. 2024ReEvo: Large Language Models as Hyper-Heuristics with Reflective Evolutioncited by 276
  6. 2025Small Language Models are the Future of Agentic AIcited by 270
  7. 2024Evolution of Heuristics: Towards Efficient Automatic Algorithm Design Using Large Language Modelcited by 268
  8. 2023EvoPrompt: Connecting LLMs with Evolutionary Algorithms Yields Powerful Prompt Optimizerscited by 193
  9. 2025ShinkaEvolve: Towards Open-Ended And Sample-Efficient Program Evolutioncited by 174
  10. 2024Meta-Prompting: Enhancing Language Models with Task-Agnostic Scaffoldingcited by 117
  11. 2024LLaMEA: A Large Language Model Evolutionary Algorithm for Automatically Generating Metaheuristicscited by 97
  12. 2016Using Centroidal Voronoi Tessellations to Scale Up the Multi-dimensional Archive of Phenotypic Elites Algorithmcited by 66
4 more
  1. 2025Barbarians at the Gate: How AI is Upending Systems Researchcited by 54
  2. 2025EoH-S: Evolution of Heuristic Set using LLMs for Automated Heuristic Designcited by 36
  3. 2024How Susceptible are LLMs to Influence in Prompts?cited by 33
  4. 2025Can Language Models Discover Scaling Laws?cited by 13

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 44.

44 papers.

Sort:

By area:

  1. 2026EvoX: Meta-Evolution for Automated Discoverycited by 60
  2. 2026Structured Scaling of AI Discovery Across Diverse Scientific Domainscited by 21
  3. 2026Dream-RSI: Recursive Self-Improvement through Evolving Worldscited by 18
  4. 2026MLS-Bench: A Holistic and Rigorous Assessment of AI Systems on Building Better AIcited by 17
  5. 2026Effective Harness Engineering for Algorithm Discovery with Coding Agentscited by 6
  6. 2026Evolution Fine-Tuning: Learning to Discover Across 371 Optimization Taskscited by 6
  7. 2026SwarmResearch: Orchestrating Coding Agents for Open-Ended Discoverycited by 6
  8. 2026Automated Discovery Has No Universally Superior Harnesscited by 6
  9. 2026PACEvolve++: Improving Test-time Learning for Evolutionary Search Agentscited by 4
  10. 2026GEAR: Genetic AutoResearch for Agentic Code Evolutioncited by 4
  11. 2026Compute Allocation for Self-Evolving LLMs: From Depth-Breadth to Multi-Armed Banditscited by 4
  12. 2026TurboEvolve: Towards Fast and Robust LLM-Driven Program Evolutioncited by 3
32 more
  1. 2026SeaEvo: Advancing Algorithm Discovery with Strategy Space Evolutioncited by 3
  2. 2026FlashEvolve: Accelerating Agent Self-Evolution with Asynchronous Stage Orchestrationcited by 3
  3. 2026What Do Evolutionary Coding Agents Evolve?cited by 3
  4. 2026Agentic Architect: An Agentic AI Framework for Architecture Design Exploration and Optimizationcited by 2
  5. 2026Open-Ended Task Discovery via Bayesian Optimizationcited by 2
  6. 2026Evolving Ensemble of Agentscited by 2
  7. 2026Heuresis: Search Strategies for Autonomous AI Research Agents Across Quality, Diversity and Noveltycited by 2
  8. 2026Knowledge-Centric Self-Improvementcited by 2
  9. 2026Relay, Don't Route: Adaptive Population Handoff for Cost-Efficient LLM-Driven Evolutioncited by 2
  10. 2026Improved Upper Bounds for Slicing the Hypercubecited by 1
  11. 2026Autopoiesis: A Self-Evolving System Paradigm for LLM Serving Under Runtime Dynamicscited by 1
  12. 2026From I/O to Code with Discovery Agentcited by 1
  13. 2026Evolutionary Intelligence for Scientific Discovery: From Evolutionary Computation to Cumulative Discovery Systemscited by 1
  14. 2026The Optimizer Is the Agent: Reasoning-Driven Search across Prompts, Programs, and ML Workflowscited by 1
  15. 2026Stateless Language Agents: Scaling Long-Horizon Automated Researchcited by 1
  16. 2026ParEVO: Synthesizing Code for Irregular Data: High-Performance Parallelism through Agentic Evolutioncited by 0
  17. 2026Evolutionary Search for Automated Design of Uncertainty Quantification Methodscited by 0
  18. 2026TuxBot: Semantic-Aware Online OS Tuning with Large Language Modelscited by 0
  19. 2026Algorithmic algorithm development with LLMs: A Case Study on LLM-Usage for Contraction Order Optimization in Tensor Networkscited by 0
  20. 2026Code evolution for link prediction in complex networkscited by 0
  21. 2026Rehearse: Stepping Back from the Confidence Cliff in Self-Improving Autoresearchcited by 0
  22. 2026varepsilon-MemEvo: Adaptive Cross-Task Memory Transfer for LLM Program Evolutioncited by 0
  23. 2026Loreley: Repository-Scale Program Evolution with Quality-Diversity Searchcited by 0
  24. 2026Test-Time Scaling for Scientific Equation Discoverycited by 0
  25. 2026iCoder-27B: Recursive AI-Led Development of Frontier Industrial Coding Modelcited by 0
  26. 2026Beyond Scripted Search: Sample-Efficient Reward Discovery via Agentic Black-box Optimizationcited by 0
  27. 2026HeurEvo: Agentic Evolution of Hybrid Solver-Augmented Heuristics for Time-Critical Mathematical Optimizationcited by 0
  28. 2026EvoDuet: Bilevel Co-Evolution of Web Searching and Task Solving for Scientific Discoverycited by 0
  29. 2026LabBook: Harnessing Experimental History for Efficient LLM-Driven Discoverycited by 0
  30. 2026FrugalEvo: Towards Cost-Aware LLM-Guided Program Evolutioncited by 0
  31. 2026CURIO: Curiosity-Driven Test-Time Learning for Open-Ended Discoverycited by 0
  32. 2026EvoAlloc: A Self-Evolving Resource Allocation Agent for Efficient Program Evolutioncited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025LoongFlow: Directed Evolutionary Search via a Cognitive Plan-Execute-Summarize Paradigm
  2. 2026SEDIMA: Cross-Run Hierarchical Insight Memory for Evolutionary Search Agents
  3. 2026CliffSearch: Structured Agentic Co-Evolution over Theory and Code for Scientific Algorithm Discovery
  4. 2026FreeEvolve: Learning to Evolve Beyond Fixed Loops
  5. 2025Evolution without an Oracle: Driving Effective Evolution with LLM Judges
  6. 2026Simple Baselines are Competitive with Code Evolution
  7. 2025Evolutionary thoughts: integration of large language models and evolutionary algorithms
  8. 2026AlgoEvo: Self-Evolving Agentic Search for Automated Algorithm Discovery