Memo

Overview

Memo is a home robot developed by Sunday Robotics designed to autonomously perform household tasks such as clearing tables, loading dishwashers, and folding laundry. It uses AI models trained on real human task data collected via a Skill Capture Glove. Memo moves on a wheeled base with a vertical reach system.

Memo runs on a wheeled base rather than legs — a deliberate design choice Sunday attributes to stability, energy efficiency, and passive safety. According to Sunday, if power is cut in any state, the robot does not fall. A telescoping Z-axis spine allows it to reach from floor level to 2.1 meters. The end-effector is a custom dual gripper, not a five-fingered hand. The exterior is soft silicone with no sharp corners. Battery runtime is four hours on a single charge. Sunday has not publicly disclosed arm DoF, compute specifications, or sensor details.

Memo is trained on ACT-1, Sunday's in-house foundation model, built from approximately 10 million household task episodes collected across more than 500 real homes. The data is gathered via Sunday's patented Skill Capture Glove — a wearable costing roughly $200 to produce that records motion and force data as people perform chores. Sunday says its Skill Transform pipeline converts that human data into robot-compatible movements with a claimed 90% success rate. The company explicitly avoids teleoperation as a training method, arguing it is too slow and capital-intensive to scale.

Sunday was founded by Stanford PhD roboticists Tony Zhao (CEO) and Cheng Chi (CTO), whose prior academic work includes ALOHA and diffusion policy research. The company emerged from stealth in November 2025 with $35 million from Benchmark and Conviction. A 'Founding Family Beta' of 50 households is planned for late 2026. Sunday states current build cost is approximately $20,000 per unit and has publicly targeted a future retail price under $10,000. No confirmed pricing or release date exists as of mid-2026.

Specifications

Sensors

Software