Plandex

Plandex is the open-source terminal-based AI coding agent designed specifically for handling large projects, large files, and large tasks — addressing limitations that other AI coding tools face when codebases exceed traditional context windows. Released as open source, Plandex features a diff review sandbox, configurable autonomy modes (from full auto to granular step-by-step), and smart context management that supports up to 2M tokens of project context. Pricing follows the BYOK (Bring Your Own Key) open-source model: Plandex itself is free, with users paying only for the LLM API usage they configure. Plandex Cloud (the company's hosted offering) is winding down per public communications, leaving the project as a self-hosted, self-managed open-source tool. Users typically run Plandex against Claude, OpenAI GPT, or local models, with API costs depending on usage intensity and chosen model — typical workflows cost $10-100/month for serious development use. Plandex's differentiation versus Aider, Claude Code, and Cursor is the explicit focus on large-task handling: rather than optimizing for quick single-file edits, Plandex is designed to plan and execute multi-hour development tasks across many files, with diff review checkpoints that let developers approve incremental changes rather than reviewing massive end-state diffs. The 2M token context management is meaningfully larger than typical IDE-embedded tools and supports working on large codebases without manual context curation. The platform combines best-of multiple LLM providers (the "Combine The Best Models" approach) by routing different sub-tasks to different models — using GPT for one type of analysis, Claude for another, local models for cost-sensitive operations. Plandex is open-source on GitHub with active community contribution. The platform is suited for power users comfortable with terminal workflows and large-task development cycles.