Advertisement:
Treehuggersupply
swisswaffen-blog

Topaz Photo Ai 3.5.0 Verified -

Additionally, the "Reveal" engine is currently limited to 480x480 pixel patches. In v4.0, expect full-frame focus recovery.

Looking at the code structure of 3.5.0, eagle-eyed users have found references to "Video Frame Stacking." It appears Topaz is preparing to bring Photo AI’s denoising architecture to video clips (currently handled by Topaz Video AI, which is less refined for photography). We may see video-to-gif stabilization directly within Photo AI by the end of 2025. Topaz Photo AI 3.5.0

In February 2025, Topaz Labs released Photo AI 3.5.0 , a significant update that introduced cloud-based processing and critical recovery features to the "all-in-one" photo enhancement tool. The Evolution of Version 3.5.0 Additionally, the "Reveal" engine is currently limited to

Users will notice a tangible reduction in preview latency. In previous versions, zooming in to 100% to check noise reduction could result in a few seconds of loading time. In 3.5.0, the caching system has been overhauled, allowing for near-instant feedback when adjusting sliders. This fluidity is essential for culling through hundreds of images from an event shoot. We may see video-to-gif stabilization directly within Photo

In version 3.5.0, the Autopilot logic has received significant under-the-hood updates. The AI is now better at distinguishing between noise and texture. Previously, aggressive denoising could sometimes smooth out skin tones or remove the granular detail of a bird’s feathers. The 3.5.0 update introduces more refined recognition protocols. It can identify subjects—such as faces or animals—and apply different levels of processing to the subject versus the background. This results in a more natural, "optical" look rather than the plastic, over-processed look that plagued earlier AI editors.

swisswaffen-blog