Sublinear

Enabling Robot Evaluation and Learning at Scale

Get API keyView documentation

Rewards in action

One API for evaluating robot behavior

Goal: insert and lock the portafilter into the espresso machine

Goal: put all the dishes into the sink

Goal: open the door and go through the doorway

Goal: use the peeler to peel the skin mandarin

Available models

Choose a model

Run every model through the same API and switch between them with a single parameter.

SOLE-R1

sole-r1

Robometer

robometer

TOPReward

topreward

RoboReward

roboreward

Python quickstart

Create an API key and submit robot video directly from Python.

Get your API key
quickstart.py
1from sublinear import Sublinear
2

3client = Sublinear(api_key=API_KEY)
4

5response = client.responses.create(
6    model="sole-r1", 
7    task_description="Fold the clothes.", 
8    video_path="episode.mp4"
9)
10

11print(response.rewards)
12print(response.success)

Get started

Create an account, generate an API key, and run your first job.

Get started