Windows-native AI media toolkit that runs fully offline

What it does APT is a portable Windows GUI that wraps local AI models for batch image, video, and audio processing alongside an offline chat interface. It handles upscaling, watermark removal, background matting, face restoration, colorization, frame interpolation, and style transfer for media, plus vocal separation and noise reduction for audio. The chat module runs local models including DeepSeek, Phi, and Qwen—though the README confusingly labels one as “ChatGPT,” which normally refers to OpenAI’s online service rather than a local model.

The interesting bit The entire stack is built to run air-gapped: models execute locally, data never leaves the machine, and the app ships as a single zip with no installer. That is a rare stance for a C# media-processing GUI, most of which phone home or lean on cloud APIs.

Key highlights

• Runs fully offline with local inference for chat, image, video, and audio tasks
• Portable Windows executable for Windows 10 x64 1803+; no installation required
• Batch image and video processing: super-resolution, background removal, watermark erasure, face restoration, and cartoon/comic style transfer
• Audio vocal splitting and denoising; text-to-speech and speech-to-text are marked as under testing
• Multi-language support via internationalization

Caveats

• Windows-only for now; Android, iOS, macOS, and Linux are on the roadmap but not yet released
• Two audio features (text-to-speech and speech-to-text) are explicitly flagged as under testing
• A “video organization” feature that decrypts and synthesizes “downloaded cached video files” is vague and its purpose is unclear from the README

Verdict Worth a look if you want a single, offline Windows app for casual AI media cleanup and local LLM chat without cloud dependencies. Skip it if you are on macOS or Linux, or if you need a well-documented API or automation pipeline rather than a GUI.