InternVLA-M1: A Spatially Guided Vision-Language-Action Framework for Generalist Robot Policy

Benchmark Model Rank Results
robot-manipulation-on-simplerenv-widow-xInternVLA-M1#18Average: 0.717
robot-manipulation-on-libero-spatialInternVLA-M1 (LIBERO-Spatial)#53Success Rate: 98.0
robot-manipulation-on-libero-objectInternVLA-M1 (LIBERO-Object)#55Success Rate: 99.0
robot-manipulation-on-libero-longInternVLA-M1 (LIBERO-Long)#74Success Rate: 92.6
robot-manipulation-on-libero-goalInternVLA-M1 (LIBERO-Goal)#86Success Rate: 93.8
robot-manipulation-on-liberoInternVLA-M1#98Success Rate: 95.9