Software development

Droid · Factory

Semi-autonomous

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

GitHubGitLabSlackLinear/JiraCLI/SDK

Sources

Last verified: . Prices and claims move fast in this space: before signing, always re-check the official source.