Does OpenVLA run on Franka Emika Panda?
works with caveats
Capabilities
| Capability | State | Notes | Evidence |
|---|---|---|---|
| Train / fine-tune | works | ev-openvla-paper-franka | |
| Real-robot inference | works | ev-openvla-paper-franka, ev-openvla-paper-franka-tasks | |
| Simulation inference | no verified evidence | — | |
| Teleop + record | no verified evidence | — |
Caveats
- Panda results come from fine-tuning (10-150 demonstrations per task), not out-of-the-box use. The OpenVLA repo ships no Franka controller or deployment script.
- OpenVLA takes a single image and predicts single-step actions. The Franka setups ran 5 Hz (Franka-Tabletop) and 15 Hz (Franka-DROID) non-blocking controllers.
Evidence 2 sources; the cell shows the highest tier
- T1 · upstream documentedpapersource 2024-09-05verified 2026-10-24 (1d ago)
OpenVLA paper, Sec. 5.2: OpenVLA is fully fine-tuned and evaluated on a table-mounted Franka Emika Panda (Franka-Tabletop) and on the DROID Franka setup.
“We test OpenVLA in two setups: Franka-Tabletop, a stationary, table-mounted Franka Emika Panda 7-DoF robot arm; and Franka-DROID, the Franka robot arm setup from the recently released DROID dataset [11],”
https://arxiv.org/abs/2406.09246v3 - T1 · upstream documentedpapersource 2024-09-05verified 2026-10-24 (1d ago)
OpenVLA paper, Appendix B.3: six Franka-Tabletop (Panda) tasks and one Franka-DROID task, 129 rollouts per approach.
“We collect 10–150 demonstrations of each of seven tasks. The first six tasks correspond to a robot setup which we denote as “Franka-Tabletop” (Franka Emika Panda robot mounted on top of a table),”
https://arxiv.org/abs/2406.09246v3
Embed this cell
Badge for READMEs and model cards. It always shows the current state.
[](https://caniembody.pages.dev/compat/franka-panda/openvla)<iframe src="https://caniembody.pages.dev/embed/franka-panda/openvla" width="380" height="132" style="border:0" title="OpenVLA on Franka Emika Panda | CanIEmbody"></iframe>History
- 2026-10-25 fecccd1 Initial release: robot policy compatibility matrix · Danilo Znamerovszkij