Agents / Coding Agent
A paper page about FrogNano, a 4B coding agent trained with RL on roughly 1,500 SWE environments using synthetic tasks and no distillation from a larger model.
The page clearly describes an AI coding agent research project, explains the training approach, and gives enough visible detail to be useful in a public directory listing. It identifies the model size, core method, intended use case, and the paper’s main claims.
This is a paper summary page, not a product or repo page, so visitors cannot tell whether FrogNano is available to try, where code or weights live, or what the exact evaluation results and benchmarks were beyond the curator’s summary.
Last evaluated 2 days ago. Current rank #1910. Down 1 spot in the rankings.
An autonomous AI agent that forecasts station temperatures and trades weather markets on Kalshi and Polymarket, with paper trading, risk limits, and public performance tracking.
