Myungkyu/rldx_1_robodojo_preset_luna_b64_60k

rldx1robodojopresetlunab6460k

Low-level policy for RoboDojo long-horizon (8 real-robot bimanual tabletop tasks, 100 demonstrations each), fine-tuned from RLWRLD/RLDX-1-PT on `Myungkyu/RoboDojo-preset-luna` — demonstrations with dense subtask labels from the subtask preset.

• —Architecture: RLDX-1-PT (video length 4, three live camera views, no keyframe slot)
• —Optimizer batch 64, 60000 steps, final checkpoint
• —Inputs: head + left/right wrist images, proprioception, the current subtask text (no keyframe input)
• —Training subtask labels annotated by GPT-5.6 Luna with the preset (Baseline) context (dataset RoboDojo-preset-luna); evaluate with the same planner model

Configs reference the base backbone / tokenizer by hub id or by the training site's local path — point them at your local copies before loading.