# Agility Robotics

> Agility Robotics builds Digit, a humanoid robot purpose-built for warehouse logistics — already deployed at GXO and Spanx, with a warehouse-first strategy distinct from general-purpose competitors.

Source: https://metavert.io/agility-robotics  
Updated: 2026-03-22

Agentic Economy

[View Market Map](https://metavert.io/agentic-market-map)

Layer 7: Physical Infrastructure as Digit

**Agility Robotics** is a humanoid robotics company building Digit, a 5'9", 140 lb bipedal robot purpose-built for [warehouse and logistics](https://metavert.io/warehouse-and-logistics-robotics) applications. Founded in 2015 as a spin-out from Oregon State University's Dynamic Robotics Laboratory, Agility has raised $683 million at a $1.75 billion valuation and represents the "warehouse-first" strategy in the [humanoid robot](https://metavert.io/humanoid-robots) race — choosing narrow commercial deployment over general-purpose ambition.

## Digit

Digit is designed to work alongside humans in existing warehouse infrastructure. At 5'9" and 140 lbs with a 35 lb payload capacity, Digit is sized to navigate standard warehouse aisles, reach standard shelving heights, and carry standard tote weights. The robot handles picking, packing, tote movement, and machine tending — tasks that are physically demanding, repetitive, and difficult to staff. Digit is already deployed at GXO Logistics (the world's largest pure-play contract logistics provider) and Spanx, with additional deployments ongoing.

Agility published details of a whole-body control foundation model for Digit, trained via [sim-to-real transfer](https://metavert.io/sim-to-real-transfer), that enables coordinated locomotion and manipulation — walking while carrying objects, adjusting gait for different payloads, and maintaining balance during reaching and lifting. This marks the transition from scripted warehouse movements to learned, adaptive behavior.

## Strategic Position

Agility's warehouse-first strategy bets that the path to humanoid robot commercialization runs through proven value in a specific domain before expanding to generality. This contrasts with [Figure AI's](https://metavert.io/figure-ai) and [Tesla's](https://metavert.io/tesla) approach of pursuing general-purpose capability from the start. The warehouse bet has advantages: the environment is controlled (flat floors, known layouts, consistent lighting), the tasks are well-defined (move tote from A to B), and the economic case is clear ($15–25/hour human cost vs. $3–5/hour robot operating cost). Agility can generate revenue and iterate on reliability in production while competitors are still in pilot programs.

The risk: if general-purpose VLA-powered robots from Figure or Tesla achieve warehouse capability as a subset of broader capability, Agility's niche advantage narrows. The counter-argument: warehouse logistics is hard enough that purpose-built optimization will outperform general-purpose robots in this domain for years, just as purpose-built industrial robots outperform humanoids at welding and painting despite humanoids being more "general."

## Related Topics

- [Humanoid Robots](https://metavert.io/humanoid-robots) — Digit
- [Warehouse & Logistics Robotics](https://metavert.io/warehouse-and-logistics-robotics) — Primary deployment domain
- [Sim-to-Real Transfer](https://metavert.io/sim-to-real-transfer) — Training approach
- [Robotics](https://metavert.io/robotics) — Parent domain
- [Figure AI](https://metavert.io/figure-ai) — Competitor (general-purpose)
- [Boston Dynamics](https://metavert.io/boston-dynamics) — Competitor (legacy)

## Further Reading

- [Agility Robotics](https://www.agilityrobotics.com)
- [Training a Whole-Body Control Foundation Model](https://www.agilityrobotics.com/content/training-a-whole-body-control-foundation-model) — Agility Robotics