Skip to main content
A modern VLA model starts broad and general. Fine-tuning is how you specialize it: take a capable base model and adapt it to your robot, your scenes, and your tasks using trajectories you’ve collected. It’s almost always cheaper and more reliable than training a policy from scratch.

​
Why not train from scratch?

Training a large policy from zero needs enormous data and compute, and throws away everything a pretrained model already knows about perception and motion. Fine-tuning keeps that foundation and only teaches the model what’s new about your task — far less data, far less time.

​
Full fine-tuning vs LoRA

The naive approach updates every weight in the model. That’s expensive to train, expensive to store, and gives you one monolithic model per task. LoRA (Low-Rank Adaptation) takes a smarter route: freeze the base model and train small adapter weights that slot on top of it.
Full fine-tuningLoRA
Params trainedAll of themA tiny fraction
Training costHighLow
StorageA full model per taskOne base + small adapters
Swapping tasksReload a whole modelSwap an adapter
Because adapters are small and composable, you keep one base model and attach task-specific adapters as needed — cheap to train, fast to iterate, easy to ship.

​
Where the training data comes from

Fine-tuning needs trajectories, and good trajectories are the hard part — see Physical AI at a glance. Cadenza closes that gap by generating data as a byproduct of simulation: every mission is logged tick-by-tick and scored automatically, so rollouts arrive already labeled and ready to become a training set.

​
The Cadenza workflow

The full loop spans both layers of the stack — you build in the SDK and operate with the CLI:
  1. Build the robot, scene, and policy interface in the SDK.
  2. Gather scored rollouts by running missions in simulation.
  3. Convert those rollouts into VLA training data.
  4. Train a LoRA adapter on that data.
  5. Evaluate the new policy back in simulation — and on hardware.
  6. Repeat, each pass producing a sharper policy and more data.
From the CLI, steps 2–4 are a couple of commands: 
cadenza env run my-project --headless   # gather scored rollouts
cadenza env finetune my-project         # rollouts -> VLA training data
cadenza env lora my-project             # train a LoRA adapter

Build with the SDK

Stand up a robot and a scene to gather data from.

Run the loop in the CLI

Gather rollouts and fine-tune end to end.