LEGO-RL: Harness-Native Reinforcement Learning for Coding Agents
Paper • 2608.17393 • Published • 25
Lego-RL
Harness-native RL for coding agents: the trained policy and the training task index.
-
-
Lego-X/qwen3_5_35b_a3b_ohsdk_200k_rl
Text Generation • 36B • Updated • 738 • 4 -
Lego-X/qwen3_5_35b_a3b_cc_200k_rl
Text Generation • 36B • Updated • 531 • 1 -
Lego-X/qwen3_5_35b_a3b_oc_200k_rl
Text Generation • 36B • Updated • 531 • 1 -
Lego-X/Lego-RL-2699
Viewer • Updated • 2.7k • 1.01k • 2
Lego-X/Lego-RL-SWE-Bench-Verified
Viewer • Updated • 500 • 162Note Evaluation set: the 500 SWE-bench Verified instances as harbor RL environments.