| Level | Method | Complexity | When to Use | Model Weight Changes | Training GPU Required |
|---|---|---|---|---|---|
| 1 | Lens Customization | Low | Adjusting behavior through configuration | No | No |
| 2 | N-Shot Examples | Medium | Guiding output with labeled examples at inference time | No | No |
| 3 | Fine-Tuning | High | Training a new model instance on your dataset | Yes | Yes |
Customization
Overview
Newton is a foundation model for reasoning about and acting on the physical world. You can customize Newton in three primary ways, progressing from lightweight configuration to deeper adaptation using your own data.
There are three ways to customize Newton, ranging from basic lens configuration to training a custom model on your data:
These customization options work together, letting you apply one or several to tailor Newton to your specific needs. Each approach varies in complexity and cost, so we suggest beginning with lens customization and advancing to more complex methods only as your requirements evolve.