The Complete Google Flow Course: Beginner to Pro

Compare Flow's three image models — Nano Banana Pro, Nano Banana 2, and Lite — including quality differences, generation limits per cycle, aspect ratios, and how credits work for AI image creation.

Learn to prompt Nano Banana step by step, from simple scenes to detailed subjects, styles, and camera direction, plus how to rename, reuse, favorite, and download your AI-generated images.

A practical prompt template for consistent AI images, covering image type, subject, action, setting, lighting, mood, and camera angle, with real examples from simple to highly detailed prompts.

Turn weak prompts into strong ones using Gemini, GPT, or Flow's built-in Agent. Learn the exact instructions to get a polished, detailed prompt without spending an hour writing it yourself.

Learn how to edit AI-generated images with simple text prompts, manage version history, save edits as new copies, and understand how selected versions appear in your workspace.

Master Flow's draw tools, Brush, Box, and Annotations, to precisely edit AI images through doodles, inpainting, and outpainting, with real prompt examples for each technique.

Learn how the Lasso and Crop tools differ from drawing tools in Flow: selecting areas without editing pixels, reframing ratios, and why cropping can reduce image quality.

Learn how to use Flow's Media Panel to find, filter, and reuse assets across projects, the difference between adding a reference and locking an ingredient with @, and why naming assets matters.

Learn how to combine multiple reference images as ingredients to build new AI images, lock specific elements with @, and understand Flow's 10-ingredient limit for consistent results.

Understand the real difference between using an image as a loose reference and locking it as an ingredient, with side-by-side tests showing how each affects style, subject, and consistency.

Build multi-angle character reference sheets for maximum consistency, extract individual grid cells, edit a whole sheet at once, and combine multiple characters into one AI-generated scene.

More hands-on examples of building consistent reference sheets in Flow, including how to adapt the character grid template for products like a Gibson Les Paul guitar.

Learn how to use Gemini to analyze videos and images you like, extract camera angles, shots, and styles, and turn them into reusable AI prompts to build your own visual vocabulary.

Get Gemini to break down any reference image into subject, pose, setting, and framing descriptions, then apply that professional composition language to your own AI-generated characters.

Create reusable AI characters with a unique visual identity and voice, fine-tune reference images, generate custom voice profiles, and use characters consistently across images and videos.

Learn how Flow's Avatar feature turns a selfie video and voice recording into your own AI likeness, how privacy works, and how to use your avatar in AI-generated images and videos.

Learn how to manage growing AI projects using Flow's categories, collections, and folders, including renaming, downloading, and safely deleting assets versus entire projects.

A practical walkthrough of Flow's search, filters, grid and batch views, and key settings like Sound on Hover, Return Silent, and Clear Prompt on Submit, to customize your workflow.

Compare Flow's video models — VEO 3.1 (Lite, Fast, Quality) and Omni Flash — including cost per generation, clip length, aspect ratios, and which model to pick for cinematic shots versus editable, multimodal video.

Learn how text-to-video works in Google Flow with no reference images, why it's the least predictable generation method, and when to use it for quick idea testing versus Frames or Ingredients for consistency.

Learn how to extend an existing AI video from its last frame, understand the VEO vs Omni Flash editing and extending limits, and manage extended clips as Scenes or standalone saved videos.

A practical video prompting template covering style, composition, subject, action, setting, lighting, camera motion, and dialogue, with real examples showing what AI video models handle well and where they struggle.

Learn why Frames to Video is the most controllable way to generate AI video, how to set start and end frames with Nano Banana images, and best practices for smooth, consistent transitions.

Master reliable camera techniques for Frames to Video, including zoom in and out, static establishing shots, and dolly or push-in motion, plus how to avoid frame conflicts that break a generation.

Learn how the Ingredients method creates consistent AI video subjects using locked references, how many ingredients VEO and Omni Flash support, and when to build subjects in Nano Banana first.

Learn how to combine several reference ingredients, like a subject, clothing, and props, into one AI video, plus the tradeoffs between locking settings, UGC-style prompting, and using Frames instead.

Learn how Flow's Scenes feature replaced Scenebuilder, letting you align multiple video clips on one timeline, trim and reorder them, save frames as assets, and export a full sequence.

Learn how to edit existing AI videos using Omni Flash prompts, including swapping objects and clothing with reference images, changing locations, and adjusting camera position and movement.

Learn how to write multishot prompts with timestamps to generate several distinct camera shots in a single AI video generation, plus tips on pacing and shot length for realistic results.

Learn advanced multishot techniques using storyboard grids and standalone reference ingredients to guide each shot, and how these compare to Frames to Video for consistency and cost.

Learn the Flow Agent's core interface — sessions, generation settings, and the 'ask before creating' safeguard — plus how to brainstorm ideas conversationally and refine weak prompts into polished ones.

Use the Flow Agent to generate a full script and a visual storyboard grid from one idea, then combine characters and image references to build out an entire AI-generated story.

Learn how the Flow Agent remembers your recent generations for fast iterative edits on images and video, and why phrasing like 'the last generated video' keeps edits accurate.

Learn how to generate multiple image or video variations in one request, batch-edit entire collections at once, and speed up product or lifestyle shoot workflows with the Agent.

Set persistent Agent Instructions to lock a specific visual style, like a Wes Anderson look, across an entire project, so every generation follows the rules without repeating them.

Use the Flow Agent to rename, trash, and organize assets into collections with simple commands

An overview of Flow's AI Tools suite for images, video, and prompting, including experimental and custom tools, and why a paid subscription is required to access them.

Learn Flow's Draw tool for sketch-based image generation with custom style presets, and Scene Explorer for brainstorming location ideas and camera angles from a simple prompt.

Combine base images with built-in mockup presets, then use Flow's layered Image Editor for inpainting, outpainting, cropping, background cutout, and AI-powered layer blending.

Generate varied camera shots from one reference image with Shot Explorer, and use Mask Magic's smart selection and brush tools to precisely inpaint or remove elements.

Learn Converge for AI retexturing with vector layers, and Grid Architect for building automated storyboard grids using Unified, Sequence, or Wizard generation modes.

Learn how to remix an existing Flow tool or build a brand-new one from scratch by chatting with the Agent, and how Gemini can help you write a strong tool-creation prompt.

Learn Flow's Shader FX for color grading, filters, and finishing effects on video, and the Type Overlays tool for adding animated text layers without leaving Flow.

Stack glitch and color effects with Pixel Bento's recordable automation, and build layered text, media, and solid-color compositions with the Poster Designer tool.

Animate hand-drawn sketches, generate AI-powered transitions between two video clips, and stack glitch effects with Weirdcore for experimental, reorderable results.

Reframe videos to a new aspect ratio, sequence multiple clips on a simple timeline with Stringout Creator, and experiment with glitchy granular synthesis effects on video.

Get AI feedback on a character's wardrobe, style, and lore with Character X-Ray, and generate a detailed written style guide from reference images with Style Writer.

Walk through Flow's Storyboard Studio: writing or pasting a script, auto-filling characters, locations, and props, and generating a full shot-by-shot visual storyboard.

Build organized, layered prompts with Prompt Tree, and create sequential shot-by-shot storyboards with Story Sketch by generating each new shot from the previous one.

Learn how Omni Flash can overlay hand-drawn doodle animations directly on top of existing AI videos, skipping After Effects or DaVinci Resolve for simple motion graphics work.

Write style and interaction prompts to generate synchronized 2D doodle animations, like shockwaves, ribbons, and impact lines, that react to a subject's exact movement and physics.

Use the Flow Agent or Gemini to generate new doodle VFX ideas from a reference video, then swap simple doodles for 3D wireframes, smoke, or glitch-style animated overlays.

Use Gemini to research and describe new visual effect styles, then bring those descriptions into the Flow Agent to generate fresh doodle and VFX animation variations.

Learn how to sketch a doodle reference on a video frame, describe it with the Agent, and use timestamped prompts to control exactly when each VFX element appears on screen.

Learn how to create stop-motion-style collage transitions in Omni, from writing transition-in and transition-out prompts to generating the paper-cutout style reference images with Nano Banana.

Build an infinite zoom-out, stop-motion paper collage video in Omni, using core directive, art direction, and kinetic text supers to animate a multi-scene story with on-screen titles.

An optional hands-on practice project applying everything learned so far: creating a consistent AI subject and placing them in different outfits and scenes for an urban fashion promotion video.

Learn how to create a clean starting frame for Frames to Video, then use Gemini with a structured prompt to isolate and describe a pose for use in an AI image generator like Nano Banana.

Learn how to refine Gemini's pose descriptions, separate subject, pose, and background details, and build a second consistent scene with new clothing and accessories.

Learn how to ask Gemini for a detailed framing description alongside subject, pose, and setting, then use it to build two more consistent AI fashion scenes with new outfits.

Turn the Scene 3 reference image into a cinematic Frames to Video generation, then explore variations like removing sunglasses and having the subject walk directly toward the camera.

Generate several video takes of the same scene, walking, zooming into clothing details, and a dynamic spin pose, and compare slow versus high-energy camera direction.