Google has added four new features in Flow, its AI tool for storytelling, that offer more precise control over images and videos. The upgrades include generative imaging with Nano Banana Pro, doodle prompts, an object insertion/removal tool and camera motion. Flow was introduced in May and offers the ability to edit and build scenes using natural language. The improvements aim to make Flow output more polished. “In Flow, you can use images to serve as the characters, subjects and starting points for your clips” with pictures you upload or create in Flow with the new “Images” tab, according to the company.

“This new dedicated workspace for generating and refining images is powered by Imagen and Nano Banana for free users, and subscribers also have access to Nano Banana Pro, our newest state-of-the-art image model which provides improved professional-grade controls like depth of focus, lighting and color grading,” Google explains in a blog post.

Simple prompts can be used to change a character’s outfit or pose and adjust the camera angle or lighting “without re-rolling the entire scene.” You can also blend elements from multiple reference images to frames, building on the original key ingredients.

“The latest updates make your Flow projects feel more like cinematic videos, not rough drafts,” suggests Digital Trends, noting that Flow’s Images tab “lets you create or refine characters and visuals before turning them into video,” while “Nano Banana Pro adds greater control over lighting, focus and color.”

Google Flow eliminates the need for multiple tools and “lets you bring ideas alive just with imagination and a prompt,” Digital Trends reports, pointing out it gives you “the tools to create fully cinematic videos with audio and polished editing, right inside the browser,” handling “end-to-end” creation without the need for separate editing software “for layering sound or cleaning up scenes.”

A Frames to Video feature lets you “draw and annotate an image directly in Flow” as part of the doodle prompt capability, writes BGR, which explains that “the AI will understand the instructions on the image and use them alongside the text prompt you might submit with the photo when creating the video.”

Google says it has “seen over 500 million videos created” since Flow’s launch alongside Veo 3 at Google I/O this spring. In October the tool was further enhanced with the integration of Veo 3.1.