Introducing EasyEvolve
EasyEvolve is a new project from PIE Research: the simplest LLM-driven evolution API. Write a function that scores a program, and get back a better one.
- You define "better". Write one evaluator that runs a candidate program and returns a score. An LLM proposes variations, each one runs in a sandbox and gets scored, and the best ones carry on to the next round.
- Get the best program back. You receive its code, its score, and the full history of the run.
- See costs before you start. Get an upper-bound estimate in USD before anything runs. Set a hard budget, and the search stops before going over it while keeping the best program found so far. Runs are billed at cost plus 10%.
- Choose a model level. Pick low, middle, or high to trade cost for capability.
