Native Windows AI translator with OCR, hotkeys, and CLI support
ait-native captures screen text with built-in OCR, sends it to a chosen LLM backend, and returns translated text via a global hotkey workflow. Developed by Weita for Windows, the app targets professionals and developers who need fast, local AI tasks. It offers native, non-Electron performance, support for OpenAI/Anthropic/Gemini backends, history tracking, and CLI/API hooks. Ideal for power users who want low resource use and direct automation control.
What does ait-native do in practical terms?
ait-native runs native Windows AI tasks: capture, OCR, send prompt, receive result. The app uses global hotkeys (Ctrl+Alt) to trigger captures and offers history and collection management for past translations, so users can repeat or refine tasks. It connects to cloud providers like OpenAI, Anthropic, and Google Gemini, and can also run local models via its native runner for offline processing.
Does it slow down your system during active use?
The app is a native, non-Electron implementation that only consumes resources when invoked, avoiding persistent background services. Because it targets native performance and minimal footprint, active scans and local processing remain constrained to the process lifecycle. The product requires a CPU with AVX2 support for some local processing features, which affects whether advanced local tasks can run on older hardware.
Is it safe to use and suitable for automation workflows?
ait-native supports direct API communication with cloud providers and local model execution, and the design prioritizes privacy by allowing local models. The tool exposes a command-line interface and an API for automation, enabling developer workflows and scripted tasks. Its lack of a background service means changes occur only when the user starts them, and history features let operators review prior runs before reapplying actions.
Clear fit for power users who control local and cloud AI tasks
The app is a dependable option for professionals and developers who need fast, native AI translations and programmable workflows. Its design choices favour low runtime footprint and scriptable access, while AVX2 dependency limits local model work on older machines. Recommended for desktop users who run repeatable translations or integrate AI into automated pipelines.
