Enables image generation, editing, and analysis using Google's Gemini 2.5 Flash and Gemini 3 Pro models, with support for batch processing, style templates, and high-resolution output.
Enables AI agents to inspect image metadata, evaluate print viability, and autonomously upscale images to 8K or 16K using RealESRGAN with GPU acceleration and CPU fallback.
Enables high-resolution image adaptive tiling preprocessing for the DeepSeek vision model, allowing analysis of large images via a six-stage pipeline that optimizes performance and handles degradation scenarios.