Google Flow AI Tutorial for Beginners: How to Create Your First Image and Video in 2026
Opening Google Flow AI for the first time can feel overwhelming — there is a lot there. This tutorial walks through everything step by step, from how to get in to generating your first image and video. By the end you will have made something real.
How to Access Google Flow AI
Before anything else — let us make sure you can actually get in. This takes about two minutes if you have a Google account already.
Open Google Flow AI in your browser
Go to labs.google/flow in your browser. Chrome works best. If you are using Firefox, Brave, Edge, or Safari and something looks broken, switching to Chrome usually fixes it immediately.
Sign in with your Google account
Click "Sign in" and use any Google account. Gmail accounts work. Google Workspace accounts work. The account needs to belong to someone who is 18 or older — Flow applies the same age requirement that Whisk AI did.
Accept the Google Labs terms
If this is your first time using any Google Labs tool, a terms of service page will appear. Read through it and accept to continue. This only happens once per account.
You are in — start exploring
The main Flow workspace loads. It looks more complex than Whisk AI did — and that is because it is. Do not let that slow you down. The next section explains every part of the interface clearly.
⚠️ Seeing "not available in your country"? Flow is available in 149+ countries. If you see a region error, try opening Flow in Chrome Incognito mode and clear your browser cache. Our Flow AI not working guide covers every confirmed access fix step by step.
Understanding the Interface
When you first open Google Flow AI you will see five main areas. Here is what each one does so you are not guessing.
Creation Panel
Left side of the screen. This is where you tell Flow what you want to make — Generate Image, Generate Video, Ingredients to Video, or Frames to Video.
Canvas / Preview
Center of the screen. After each generation, your result appears here. You can view it, download it, or send it to the timeline for editing.
Library
Right side or top menu. Your Library stores everything you have generated — images, video clips, ingredients, full projects. Think of it as your personal project folder inside Flow.
Timeline
Bottom of the screen. When building a longer video project, drag clips from your Library into the timeline, reorder them, and preview the full sequence.
Credits Display
Top right corner. Shows your remaining generation credits. Free accounts receive limited daily image credits. Video generation credits require a Google AI Plus subscription.
Free tier tip: Image generation has generous daily limits on the free tier. Video generation with Veo 3.1 requires a Google AI Plus subscription ($19.99/month). For your first sessions, start with image generation to get comfortable before upgrading.
Your First Image Generation — Complete Walkthrough
Let us make your first image. This is the simplest thing you can do in Flow and a good way to get comfortable with how it responds to descriptions.
Click "Generate Image" in the creation panel
Select the image generation option on the left side of the interface. A text box will appear in the center of your screen.
Write your first description
Type a description of what you want. Start simple — do not try to describe everything at once. A good first prompt:
A golden retriever puppy sitting in a sunlit garden, soft morning light, shallow depth of field, warm colours, photorealistic
Click Generate and wait 10–20 seconds
Flow sends your description to Imagen and produces the image. This typically takes 10–20 seconds. If it takes longer, check your internet connection — slow uploads extend generation time.
Review the result — download or refine
If you like it — download it or save it to your Library. If something is off, change one specific detail and regenerate. Changing everything at once makes it impossible to know what worked.
Generate 3–4 variations before changing anything
Hit generate again with the same description. Flow produces a different interpretation every time. Running 3–4 variations shows you the full range before you start tweaking the prompt.
Want the sticker or plushie style from Whisk AI? Those preset buttons do not exist in Flow — but the styles are achievable through text. Our Whisk AI prompts guide has the exact prompt language for each Whisk style that works in text-based tools.
Your First Video Clip — Step by Step
This is where Google Flow AI goes beyond what Whisk AI ever offered. Note: Video generation with Veo 3.1 requires a Google AI Plus subscription ($19.99/month). Follow these steps for your first generation.
Click "Generate Video" in the creation panel
Select the video generation option. Make sure Veo 3.1 is selected in the model picker — this is the model that generates synchronized audio alongside the video. Requires AI Plus subscription.
Write your video description — think in shots
Video prompts work best when you describe a camera shot, not just a scene. A good first video prompt:
A slow tracking shot following a tabby cat walking across a wooden windowsill, rain falling outside the window, warm amber indoor light, camera holds steady. Ambient sound: rain on glass, distant street noise.
Choose aspect ratio and click Generate
Select your aspect ratio — 16:9 for YouTube/desktop, 9:16 for TikTok/Reels, 1:1 for general social media. Then click Generate. Video generation takes 30–90 seconds.
Watch and evaluate your 8-second clip
Your clip will be up to 8 seconds long with audio. Watch it through completely before deciding to regenerate. The audio quality on first watch sometimes takes a moment to fully load.
Extend the clip for longer videos
Want more than 8 seconds? Click Extend on any clip. Flow generates additional footage from the final second of your clip, maintaining visual continuity. Chain multiple extensions for a full minute of video.
📌 No audio on your video? This almost always means Veo 3.1 was not selected, or your prompt did not include explicit audio direction. Add a dedicated audio sentence: "Ambient sound: [describe the sounds]." See our Flow AI not working guide for the full audio fix.
The Ingredients System — Character Consistency
This is one of the most powerful features in Google Flow AI — and the closest thing to Whisk AI's visual reference input system.
The Ingredients system lets you define visual assets — a specific character, a specific object, a specific style — and use them consistently across multiple generations. This was something Whisk AI could never do. Every generation in Whisk produced a slightly different interpretation of your subject. With Ingredients, your character looks the same across 50 different generations.
How to Create an Ingredient
Upload any image to your Library. Then click on that image and select "Set as Ingredient." Give it a name — something descriptive like "Main Character," "Product Shot," or "Background Style." That is it. The ingredient is now saved and reusable.
How to Use Ingredients in Generation
When writing a new image or video prompt, you will see an option to "Add Ingredient." Select the ones you want. You can also type @ in the prompt box to tag specific ingredients by name. Your prompt should reference the ingredient explicitly:
Important: Do not just upload the ingredient and expect Flow to use it automatically. Your text prompt must reference it. If you uploaded a character called "Sarah," your prompt should say "Sarah walks through the market" — not just "a woman walks through the market." The explicit reference produces dramatically more consistent results.
Character Consistency
Use the same character across dozens of generations. Same face, same outfit, different scenes and moods — all consistent.
Product References
Upload your product and use it in multiple lifestyle settings, backgrounds, and contexts without it looking different each time.
Style References
Save a visual style as an ingredient and apply it consistently across an entire series of images or video clips.
Veo 2 Only for Video
Ingredients to Video currently requires Veo 2, not Veo 3.1. Use plain backgrounds in ingredient images for best results.
Prompt Writing That Gets Better Results
The gap between average results and great ones almost always comes down to how you write the prompt. These are the four things that actually matter.
| Element | What it does | Example |
|---|---|---|
| Subject | Who or what is the main focus. Be specific — not "a dog" but "a golden retriever with floppy ears" | A golden retriever puppy, floppy ears, mud on its paws |
| Setting | Where is this happening and what does it feel like. Include weather, time of day, atmosphere | On a wooden porch in autumn, fallen orange leaves around it |
| Light | Single most important variable for mood. Completely changes the emotional feel of the output | Warm golden hour side light, long soft shadows |
| Camera / Quality | For images: quality anchor. For video: camera movement. Tells Flow how to frame the scene | Shallow depth of field, editorial photography quality |
| Audio (video only) | Separate sentence for audio direction. Veo generates audio much more reliably when you describe it explicitly | Ambient sound: autumn wind, distant birds, leaves shuffling |
Ready to go deeper? Our 50 best Google Flow AI prompts guide has tested prompts for every style — images, cinematic videos, dialogue scenes, product shots, and more. All copy-paste ready.
Common Beginner Mistakes — And How to Fix Them
These are the patterns that consistently produce weak results, and the specific adjustment that fixes each one.
Prompts That Are Too Vague
"a nice landscape"
"Rolling green hills at golden hour, long shadows, no people, wide cinematic shot"
Vague prompts produce generic results. Four specific elements change everything.
Too Many Conflicting Styles
"realistic photo, watercolor, cartoon, vintage, modern minimalist"
"watercolor illustration style, soft window light, children's book quality"
Conflicting styles produce muddy results. Pick one and commit to it.
Generating Once and Giving Up
Generate once → not perfect → give up or change everything
Generate 3–4 variations → identify what needs to change → adjust ONE element → regenerate
The first generation is never final. Variation reveals the range.
Burning Credits Too Fast
Generate the same vague prompt 15 times hoping for different results
Plan your prompt carefully → generate 3 variations → make one targeted change → repeat
Use your credits deliberately — start with image generation, upgrade for video.
Frequently Asked Questions
Ready to Try Your First Prompt?
Browse our collection of 50 tested Google Flow AI prompts — images, cinematic video, dialogue scenes, and more. All copy-paste ready to use right now.
Content on WhiskAILabs is created and reviewed by people who actively test AI image tools in real creative workflows.
Get Started free All tools with WhiskAILabs.
-
© 2026 WhiskAILabs.net Independent educational resource. Not affiliated with Google or Google Labs.