openevolve 开源项目推荐（20 个）

algorithmicsuperintelligence /openevolve

AlphaEvolve 的开源实现

shyamsaktawat /OpenAlpha_Evolve

OpenAlpha_Evolve是一个开源Python框架，其灵感来自DeepMind的AlphaEvolve等自主编码代理的突破性研究。

Optima-CityU /LLM4AD_Next

下一代自动算法设计平台，让自动化算法设计更易获取、更易使用

Protocol-zero-0 /evolution-kernel

适用于任意 git 仓库的自我改进代码：给它一个可衡量的目标，编码智能体（Claude Code 或 Aider）提出变更，内核对其进行评分，保留优胜方案并回滚失败方案。开源的 AlphaEvolve / FunSearch 风格进化循环，沙箱化运行，并带有可审计账本。

yuxuan-z19 /cudanalyst

[ICML'26] Towards Feedback-to-Plan Decisions for Self-Evolving LLM Agents in CUDA Kernel Generation

VeryMath /AI4Math-Evolving

AI4Math skills for search-based experiments, OpenEvolve workflows, and iterative program improvement.

kibrq /k-server-bench

Executable benchmark for AI-driven potential discovery for the k-server conjecture, with evaluators, metric instances, and released experiment workflows.

SammyTourani /tripwire

Correctness oracle for AI code-optimization loops: catches reward hacks with canonical, metamorphic, withheld-input, and timing checks.

hashkanna /mendel-evolve

Genetics for Algorithm Discovery. Evolve ideas, not programs: every LLM idea becomes a named switch, and knockouts measure what each is worth. New lower bounds at 17 sizes of Tao et al.'s problem 60. Built at the London AI x Science Hackathon 2026.

xs271828 /openevolve-scientist

Codex Skill and platform-neutral research CLI for reproducible, auditable OpenEvolve algorithm discovery with holdouts, ablations, budgets, and novelty audits.

MateoTTR /revolver

MLOps CLI for LLM-guided evolutionary optimization on a pluggable backend (OpenEvolve, ShinkaEvolve): versioned problems and datasets, LLM registry, reproducible runs, benchmarks

ptuls /randomize-evolve

Using Levi to optimize randomize algorithms for set membership and tasks in similar vein

tiroq /openevolve-retry-policy

Evaluation-first AI case study for evolving retry/backoff policies with local LLMs, strict QA gates, and holdout validation.

Sanexxxx777 /erdos-openevolve

Reproducing AlphaEvolve's SOTA bound on the Erdős minimum-overlap problem with an evolutionary LLM pipeline (OpenEvolve) — converges in ~10 iterations, two independent backends, one-command verifier

cyberdid /VoiceID

Research knowledge base and architecture for large-scale open-world speaker discovery, VoiceEvolve and OpenEvolve. Ukrainian documentation and detailed technical audit.

scttfrdmn /openevolve-bedrock

Amazon Bedrock as the LLM backend for OpenEvolve, with a per-call cost ledger

archie-brown3 /noema-evolve

NoemaEvolve extends LLM evolutionary program search with RL based reflection and a ReAct style coordination

ShumpZeke /OpenEvo

A production fork of OpenEvolve: the engine byte-identical to upstream, plus a real-time control plane, a provider broker with role-based routing, and an isolated agent sandbox.

wuhandong313 /ouroboros-quant

Ouroboros-Quant 衔尾蛇: Recursive self-improvement agent for quant research — OpenEvolve evolves signal/factor code, EvoAgentX evolves the researcher. 自进化量化研究流水线(数据→信号→检验→组合→归因)

karsten-s-nielsen /ruthless-efficiency

A general optimisation/search substrate: a pure hexagonal core + pluggable search strategies + pluggable compute backends.