There have been a lot of Pixel leaks today, and now we have another one. This is all about the Pixel 10 and Pixel 11’s camera features, which unsurprisingly have AI all over them.
But first, let’s start with hardware. The Pixel 11 Pro coming in 2026 is expected to sport a “next-gen” telephoto camera, one that will use its hardware in concert with AI features to bring 100x zoom for both photos and videos.
Cinematic Blur will be updated on the Pixel 11 Pro as well, with support for 4K30fps and a new “video relight” option that does what it implies – changes the lighting conditions inside videos. This is made possible by a Cinematic Rendering Engine in the Tensor G6’s image signal processor, which also reduces the power draw of recording with blur by almost 40%.
Ultra Low Light video is coming to the Pixel 11 too, and this is also referred to as Night Sight video. While Night Sight video already exists, the current implementation relies on the cloud for its processing, whereas the Pixel 11 one will work entirely on-device. The targeted ambient light level for this is around 5 to 10 lux.
Now let’s move back to 2025 and the Pixel 10. This series will be powered by the Tensor G5 SoC, which will enable “Post-capture Generative AI-based Intuitive Video Editing for the Photos app”. This could also be available for YouTube Shorts.
Speak-to-Tweak will be a new AI feature for photo editing, and Google is working on Sketch-to-Image as well, which is basically the same thing that Samsung’s Galaxy AI already offers. There’s a “Magic Mirror” feature coming too, but it’s unclear what this is, while the Tensor G5 should also be able to run Stable Diffusion-based models locally, which could be used in the Pixel Studio app.
The Tensor G5 will finally add support for 4K60fps HDR video, up from the Tensor G4’s 4K30fps HDR video.
