Greyfin
RL environments for trading
Markets built
for learning.
Trading agents need markets that react to their decisions.
The market responds.
Every action changes the environment an agent learns in.
The agent chooses Buy, Sell, or Hold from observed momentum, liquidity imbalance, and its position. Rewards measure portfolio change after trading costs and an inventory penalty. The dotted future is available only to this visualization. Manual control preserves the same market. Pause stops the environment; reduced motion starts paused.
Build with us.
For quantitative teams, market researchers, and AI labs.