← Leaderboard

Qwen-RobotManip-Context

VLA
Alibaba Qwen · Jun 16, 2026 · built on Qwen-RobotManip

Variant of Qwen-RobotManip that also conditions on the episode’s execution history (in-context policy adaptation).

Type
VLA
Parameters
—
Backbone
Qwen3.5-4B
Weights
Closed
License
–
Commercial use
Unclear
Released
Jun 16, 2026
Robots
Franka Panda (sim), Aloha-AgileX (sim)
SO-101
–
LeRobot
–

Benchmark results

4 of 13 benchmarks
#19 of 2633.8%, self-reported by the authors, not a leaderboard entry
#3 of 2492.0%, self-reported by the authors, not a leaderboard entry
RoboTwin 2.0Near saturationSimulation
  • 77.0%, self-reported by the authors, not a leaderboard entryPaper / model card ↗Jun 2026· self-reported, clean-only training· non-standard
77.0%, self-reported by the authors, not a leaderboard entrynot ranked
LIBEROSaturatedSimulation
#6 of 4699.2%, self-reported by the authors, not a leaderboard entry
Not reported on 9 benchmarks
Robustness drop: 99.2% on LIBERO → 92.0% on LIBERO-Plus. The perturbed variants show how much of the LIBERO score is memorization.

Adoption

updated Nov 17, 2026
GitHub stars
189
GitHub forks
9
Wrong or missing number? Report an error