← Leaderboard

OpenVLA

VLA
Stanford / UC Berkeley et al. · Jun 13, 2024

Open-source 7B VLA trained on Open X-Embodiment data that predicts discretized action tokens. It is a reference baseline for the LIBERO family.

Type
VLA
Parameters
7B
Backbone
Prismatic (Llama 2 7B)
Weights
Open
License
MIT
Commercial use
Allowed
Released
Jun 13, 2024
Robots
WidowX (Bridge), Google Robot, Franka
SO-101
–
LeRobot
–

Benchmark results

4 of 13 benchmarks
#24 of 2416.3%, self-reported by the authors, not a leaderboard entry
#5 of 657.5%, self-reported by the authors, not a leaderboard entry
LIBEROSaturatedSimulation
#45 of 4676.5%, self-reported by the authors, not a leaderboard entry
SimplerEnv WidowXSaturatedSimulation
#19 of 191.0%, self-reported by the authors, not a leaderboard entry
Not reported on 9 benchmarks
Robustness drop: 76.5% on LIBERO → 16.3% on LIBERO-Plus → 57.5% on LIBERO-PRO. The perturbed variants show how much of the LIBERO score is memorization.

Adoption

updated Nov 17, 2026
HF downloads · 30d
58.3K
HF downloads · all time
18.2M
HF likes
266
GitHub stars
7.3K
GitHub forks
873

Running the model

HardwarePrecisionConfigLatency / chunkStepsVRAMChunkControlSource
A100–fine-tuned, autoregressive, 1 image (LIBERO)239.6 ms––1–Feb 27, 2025 ↗
RTX 4090bf16no compilation or speculative decoding167 ms–15 GB1–Jun 13, 2024 ↗
RTX A5000int44-bit quantization333 ms–7 GB1–Jun 13, 2024 ↗

Trained on

Runs on

Wrong or missing number? Report an error

Built on OpenVLA