TMax: Advancing Reinforcement Learning

TMax is a new open RL recipe for terminal agents, featuring a large dataset of 14,600 RL environments. It brings open data recipes closer to the frontier.

One continuously updated timeline instead of dozens of separate articles. New developments are appended as the story evolves.

1. ReleaseJun 22, 2026, 03:38 PM 73%

   TMax: A Simple Recipe for Terminal Agents Released

   TMax is a new open RL recipe for terminal agents, featuring a large dataset of 14,600 RL environments. It brings open data recipes closer to the frontier.

   Read the full story →