← Leaderboard
ACoT-VLA
VLAAGIBOT · Jan 16, 2026 · built on π0.5
Action chain-of-thought VLA from AgiBot and Beihang University, built on π0.5. Checkpoints are not yet released.
- Type
- VLA
- Parameters
- —
- Backbone
- PaliGemma (π0.5)
- Weights
- Unknown
- License
- –
- Commercial use
- Unclear
- Released
- Jan 16, 2026
- Robots
- Franka Panda (sim)
- SO-101
- –
- LeRobot
- –
Benchmark results
2 of 13 benchmarksLIBERO-PlusSimulation
- Paper / model card ↗Jan 2026
#6 of 2487.5%, self-reported by the authors, not a leaderboard entry
- Paper / model card ↗Jan 2026
#17 of 4698.5%, self-reported by the authors, not a leaderboard entry
Not reported on 11 benchmarks
Robustness drop: 98.5% on LIBERO → 87.5% on LIBERO-Plus. The perturbed variants show how much of the LIBERO score is memorization.
Adoption
updated Nov 17, 2026Wrong or missing number? Report an error