Warp

Free: $0. The full terminal, Warp Agent CLI access, bring your own AI inference, limited cloud agents, 3 indexed codebases, 30 stored cloud conversations, and a shared Warp Drive capped at 10 workflows and 3 notebooks. No agent usage is included; you can buy usage at prices 20% above paid-plan rates, or connect your own OpenAI, Anthropic or Google API key, a SuperGrok subscription, or an OpenAI-compatible endpoint. Self-serve keys and endpoints are limited to individuals and organizations with 10 or fewer employees; bigger companies need Business or Enterprise.

Self-hosted: The client is open source (AGPLv3, with the warpui_core and warpui crates under MIT) and builds on macOS, Linux and Windows with ./script/bootstrap and ./script/run. The server, Warp Drive backend, hosted authentication and Oz orchestration are not in the repo and remain proprietary, so a self-built client still needs Warp's backend for Drive sync, hosted-model agents and team features. Self-hosted cloud agents and inference in your own cloud (AWS Bedrock, Google Vertex AI) are Enterprise-only. Contributions require a CLA.

Paid: Build: $20/mo, or $18/mo billed annually, with $20 of agent usage at API rates each month, full Warp Agent access, unlimited Drive and conversation storage, 40 indexed codebases, private email support. Max: $200/mo, or $180/mo billed annually; the plan card says $200 of usage (10x Build), while the comparison table still lists "$240 (18,000 credits)" from the old credit system. Business: $50 per user per month, or $45 billed annually, up to 25 seats, $20 of usage per seat, SAML SSO, team usage metrics, admin-configurable data controls, team BYO API keys and custom endpoints. Enterprise: custom; unlimited seats, shared usage pools, BYOLLM, self-hosted cloud agents, multi-admin controls. Purchased add-on usage rolls over and stays valid for 12 months.