OpenEvolve
An open-source take on AlphaEvolve: LLMs mutate programs, your evaluator scores them, and a MAP-Elites and island database keeps diverse winners.
The loop
- Proposes
Code changes in marked EVOLVE-BLOCK regions, written by an ensemble of LLMs from any compatible API
- Judged by
A user-written evaluator that scores each program; errors and other evaluator output are kept as artifacts
- Steers
Scores place programs in a MAP-Elites grid across island populations; errors and evaluator output feed the next prompt
Headline result
The project reports 2-3x speedups on GPU kernels (MLX Metal) and a circle packing for n=26 that its README calls state of the art, plus sorting and signal-processing examples. These are the project's own claims; the README gives no numeric circle-packing score.
What this does not prove
All results are project-reported and not independently replicated here. The README gives no numbers for most examples. It is an engineering framework, not a paper.
How it generalizes the harness
The evolutionary loop with its MAP-Elites and island database does the search, and the evolved program is a plain artifact usable without the model.
Classification
| What it proposes | Program |
|---|---|
| Generative model | Pluggable LLM ensemble, frozen (any OpenAI-compatible API or supported CLI) |
| Algorithmic authority | User-defined evaluator (tests, benchmarks, profilers) |
| Feedback | Score |
| What gets updated | Prompt context and population of candidates |
| When it runs | At inference time |
| Validity | Empirical (measured) |
| Evidence | Code released |
| Lineage | Derivative |