OpenWarp

OpenWarp is a community fork of the Warp terminal maintained by zerx-lab that opens up Warp's AI layer. It keeps the full Warp terminal experience — blocks, workflows, keymaps — while letting users plug in any OpenAI Chat Completions-compatible provider, customize system prompts with minijinja templates, and keep every credential local.

The official Warp client routes AI through Warp's cloud agent service. OpenWarp opens that layer entirely so users own the stack end-to-end.

• Bring Your Own Provider (BYOP) — Paste a Base URL and API key in settings. Any OpenAI Chat Completions-compatible endpoint works out of the box, including OpenAI, DeepSeek, Qwen, Groq, Together, local Ollama / LM Studio, and most proxy gateways.
• Local credential storage — API keys and authentication tokens never leave the machine. They are stored locally in the configuration file and requests go directly to the configured provider with no relay.
• Native multi-protocol support — Six native API protocols via the genai adapter, implemented natively rather than as a shim, preserving protocol-specific features like DeepSeek's reasoning_content or Claude's thinking capabilities.
• Dynamic prompt templates — A minijinja template engine renders the system prompt in real time based on the current working directory, locale, and role.
• First-class internationalization — Ships with English and Chinese locales, with community-extensible locale support.
• Upstream sync — Continuously merges from upstream Warp while layering BYOP and i18n on top.
• Familiar Warp UX — Switch models, conversations, and command suggestions with one click. The terminal experience is identical to Warp.

OpenWarp is built in Rust and distributed as source, with installers for macOS and Windows. It is dual-licensed under AGPL-3.0 and MIT (matching upstream Warp: warpui_core / warpui crates under MIT, everything else under AGPL-3.0). It is not affiliated with Warp Inc.