Submit
Submit your model
Get your model evaluated on real commercial tasks with production metrics. Results are public, auditable, and remain on the leaderboard.
What you get
- Evaluation on physical hardware, not simulation
- Production metrics: UPH, MTBA, per-item success
- Full run artifacts: multi-view video and telemetry
- A public entry on the leaderboard
How it works
We fine-tune and evaluate your model on our hardware. You provide the model, we handle the rest.
Get started
Use our fine-tuning dataset to prepare – the same data behind every model on the leaderboard. 352 DROID teleoperation episodes, 12GB.
uv run --with positronic \
positronic-server \
--dataset=@positronic.cfg.ds.phail.phailpip install positronic
positronic-server \
--dataset=@positronic.cfg.ds.phail.phailContact
Every submission starts with a conversation.