Factory Droid 1.0 Launches as a Coding Agent Tool
Factory has released Droid 1.0, a coding agent tool that works with any AI model and makes programmatic tool calling its default setting. On Factory's own Terminal-Bench 4 tests, Droid scored 65.2% with Claude Opus 5.5, at a cost of $4.73 per passed task.
Quick glossaryFactory, Terminal-Bench 4, programmatic tool calling
- Factory
- An AI startup known for building coding agents that help developers write and manage software.
- Terminal-Bench 4
- A standard test that checks how well AI agents complete tasks in a command-line computer environment.
- programmatic tool calling
- A way for an AI to write code that runs tools automatically, instead of calling each tool one step at a time.
Burning Potato score: 62%