Factory AI Releases Droid 1.0 With Programmatic Tool Calling

Factory AI introduced Droid 1.0 on its website, adding programmatic tool calling to its coding agent harness. The release uses a Script tool that lets models write JavaScript workflows and execute hundreds of tool calls within one model request. In benchmark tests Factory ran itself, the company says Droid scored higher at lower cost than rival coding agent harnesses.

The scripts run in a sandbox without direct file, network or system access. Tool results remain inside the program unless printed, while independent calls can run concurrently. Factory reports that Script mode reduced costs by 15% with Claude Opus 5.5 compared with direct tool calling.

On Terminal-Bench 4, according to Factory, Droid with GPT-6.1 Sol scored 3.5 points higher than Codex at 12% lower cost. With Claude Opus 5.5, it scored 3.0 to 7.6 points higher than Cursor, OpenCode, Devin and Claude Code, with costs up to 24% lower.

Droid 1.0 also changes how newly discovered tools are added to requests so previously processed prompts remain cached. Factory reports that a typical production request now reads 98% of its input from the prompt cache.