Droid · Factory
Development agents ("Droids") that execute software tasks end-to-end from CLI, desktop or cloud sandbox: legacy-code modernisation, debugging, migrations, tests and PRs, with parallel execution on Factory-managed "Droid Computers".
- Public price
- da $20/mese
- Pricing model
- Per seat
- AI model
- multi-modello dichiarato: GPT-5, Claude Opus/Sonnet, Gemini e modelli open-weight
- Target
- Developers
What it promises
"Droid is the #1 software development agent on Terminal-Bench" (58.75%, above Claude Code and Codex CLI), on any frontier model.
What is verifiable
More verifiable than average: Terminal-Bench is an open benchmark with a public leaderboard and reproducible tasks, though the run was executed by Factory itself (not a third-party lab). Public self-service pricing; $150M Series C at a $1.5B valuation (April 2026) reported by the press.
The real cost
Pro $20, Plus $100 (~5x usage), Max $200 (~10x); rolling usage limits; extra pay-as-you-go credits ($10 minimum); Teams and Enterprise by quote with SSO, Zero Data Retention and on-premise options.
Key integrations
Sources
- Factory — pagina prezzi ufficiale verified on 2026-07-05
- Factory — Droid #1 su Terminal-Bench verified on 2026-07-05
- Factory Docs — Terminal-Bench leaderboard verified on 2026-07-05