Free · 8 min read
AI video generation: where to start
How to generate video with AI: choosing between Google Flow, Gemini, Runway and Adobe Firefly, writing prompts for motion and camera, what's free and how not to waste credits.
- Which services generate video and what you can do for free
- A video prompt formula: shot, character, action, location, aesthetic
- How to get controllable results by starting from your own frame
- Free
- 1,504 members
- Join in 1 click
5 steps to your first AI video
- 1
Step 1
Choose a service
- 2
Step 2
Write prompts like a director
- 3
Step 3
Start from your own frame
- 4
Step 4
Iterate without burning credits
- 5
Step 5
Edit the video together and know the rules
- 1,504
- community members
- 8,915
- messages with questions since July 2024
- 9
- design games since September 2024
- 759
- people wrote in the chat
This guide is for you if you are
Never generated video
You've seen AI clips in your feed and want to try, but don't know which service to start with.
A designer who needs motion
You want to bring an illustration to life, make a landing page background or a short intro for a case study presentation.
Spending credits on chaos
You generate, but characters melt, the camera jumps and the result is different every time.
In short
- You can try for free in Google Flow (50 credits a day), Runway (125 credits once) and Adobe Firefly (a limited number of daily generations); video in the Gemini app needs a Google AI plan.
- OpenAI's Sora no longer works: the web and app experiences were discontinued on April 26, 2026.
- AI generators make short clips a few seconds long, so build a video from several clips in an editor.
- A video prompt describes motion, not just the frame: what the character does, how the camera moves, what happens around.
- Start from your own image: you control composition and style, and the prompt only describes motion.
5 steps to your first AI video
- 1
Step 1
Choose a service
Services change fast, so check their help pages before you start. As of November 2026, the picture is:
- Google Flow: Google's video studio, 18+ only and available in supported regions. Without a subscription you get 50 credits a day (unused ones don't roll over) and 3–10 second clips; during peak hours, roughly 2 PM to 5 PM UTC, video generation may be unavailable without a subscription;
- Gemini: video in the Gemini chat needs a paid Google AI plan and is 18+: Create video in the sidebar, clips of 3–10 seconds;
- Runway: the free plan gives 125 credits once: they don't expire but don't renew either, and you can't buy credits on the free plan. The set of models available is limited;
- Adobe Firefly: Video → Generate video, with a choice of Adobe and partner models; free users get a limited number of generations per day.
💡 If you're looking for OpenAI's Sora: its web and app experiences were discontinued on April 26, 2026, and the API on September 24, 2026. OpenAI recommends exporting old work via sora.chatgpt.com/sunset while that's possible.
- 2
Step 2
Write prompts like a director
For video, describing the picture isn't enough: you need to say what moves and how. Adobe recommends this formula for Firefly prompts:
- shot and camera: "a close-up shot with a slow zoom-in";
- character: who it is, what they look like, how they feel;
- action: what they do, with a dynamic verb and pace: "slowly", "suddenly";
- location: where, what weather, what environment;
- aesthetic: "cinematic, 35mm film, shallow depth of field".
💡 Adobe warns that more than four subjects often confuse the model, and Runway recommends positive phrasing: "locked camera" instead of "no camera movement". General prompt rules: How to write prompts for AI image generation.
Show your first clip
People in the community share video prompts and break down why a clip came out chaotic. Come in, it's free.
Show your clipFree. 1,504 designers are already in the community.
- 3
Step 3
Start from your own frame
The most controllable results come from image-to-video: you upload an image and the model animates it. Runway says that in this mode the prompt should describe motion, not the image's contents: the model already sees the composition, colors and style. For a designer, it's a way to animate your own illustration or a layout frame:
- in Runway, the Gen-4 model only works with an input image, which becomes the first frame of the video;
- in Google Flow you can set start and end frames and describe the transition, or add "ingredients": an image of a character or object so it looks the same across clips;
- Google recommends references of the subject or product on a plain or segmented background;
- in Gemini you can add one video and up to 5 images to a prompt.
💡 Refer to the character in general terms: Runway recommends "the subject" or "she" rather than describing again in detail what's already in the image, which can reduce motion or give unexpected results.
- 4
Step 4
Iterate without burning credits
Every generation costs credits, so learn on cheap tries. Runway recommends starting with a simple prompt that describes only the main motion, then adding one element at a time: subject motion, camera motion, scene motion, style. That way you see what changed the result. How to save:
- in Google Flow, 360p video costs a third of 720p: 1 credit per second versus 3 for the Omni Flash 1.2 model;
- Veo 3.1 Lite in Flow makes 4, 6 or 8 second clips in 720p for 10 credits for users without an Ultra plan;
- in Runway, Gen-4 Turbo costs 5 credits per second and Gen-4 costs 12, and Runway itself recommends testing in Turbo;
- in Firefly, the same seed, prompt and settings give similar clips, which helps when you change one detail;
- if a Flow generation fails or hits a safety filter, the credits are returned within an hour.
💡 Credit costs change: Google says to check the current cost in the prompt box settings before generating. In Runway, a clip's cost also depends on the model and duration.
Video coming out chaotic? Show your prompt to the community and ask what to change → - 5
Step 5
Edit the video together and know the rules
Generators give you clips a few seconds long, so the final video comes together in an editor: join the clips, add sound and titles. How to do that: Video editing from scratch and CapCut for beginners. Before publishing, check:
- artifacts: Adobe lists distorted human and animal forms in motion, unwanted objects and unintended motion among known issues;
- rights to input images: both Google and Adobe require you to upload only what you have rights to;
- AI labels: all Flow videos carry an invisible SynthID watermark, and you can add a visible one with the Visible watermarking toggle;
- ownership: under Google's terms, the company doesn't claim ownership of generated content, but you still need to read the usage terms in full.
💡 Don't use video generation to show real people in invented situations: Google prohibits creating videos to deceive, harass or harm others, and violations can cost you your account.
You're not alone with these questions
We counted what people ask about most in our community. AI tools are the fourth most popular topic, after learning, portfolios and Figma.
Messages by topic in Spaceberry Community, July 2024 — November 2026. Keyword search.
Practice with a deadline and live feedback
The best teacher is a task with a deadline that other people see. That's why the community regularly runs design games: the same brief for everyone, a deadline and a vote.
Design game · January 2026
NAOMA website redesign
The academy's home page through the eyes of an applicant, a student and a teacher. Three finalists; the community chose the winner.
AI challenge · May 2025
A landing page made by AI
Logo, copy and visuals made only with AI tools. Four entries, an anonymous vote, and participants shared their prompts.
Workshop · April 2026
Figmosha: AI in Figma
The tool's author showed how to generate interfaces in Figma and analyze a design system.
Free. 1,504 designers are already in the community.
How to join: 1 minute
- 1
Tap the button
Telegram opens with the Spaceberry community.
- 2
Join
It's free: all you need is a Telegram account.
- 3
Introduce yourself
Say where you're starting from and ask your first question.
Questions and answers
Didn't find your answer? Ask it in the community.
Can I generate AI video for free?
Yes, with limits. Google Flow gives 50 credits a day without a subscription, Runway gives 125 credits once on its free plan, and Adobe Firefly gives a limited number of free generations per day. Video generation in the Gemini app isn't available for free: you need a Google AI plan.
What happened to Sora?
OpenAI discontinued the Sora web and app experiences on April 26, 2026, and the API on September 24, 2026. If you still have work there, OpenAI recommends exporting it via sora.chatgpt.com/sunset; after any final export window, the data will be permanently deleted.
Which language should I write video prompts in?
Google Flow supports many languages but recommends English prompts for the best results and warns that quality in other languages may vary. Camera terms like close-up, slow zoom-in and handheld camera are worth learning in English either way.
Can I use AI video in commercial projects?
It depends on the service and plan, so read the terms of the specific tool. Google says it doesn't claim ownership of content generated in Flow, but use is governed by the terms of service. Adobe says Generate video in Firefly creates "commercially safe" clips, but if you pick a partner model, check its terms separately. And in every case you need rights to the images you upload.
Is the Spaceberry community paid?
No. Spaceberry Community is a free Telegram community supported by Spaceberry Foundation, a non-profit organization. All you need to join is a Telegram account. Inside are 1,504 designers who discuss each other's work, plus regular design games: the same brief for everyone, a deadline and a community vote. Conversations are in Ukrainian.
Generate your first clip
Join 1,504 designers who learn, share their work and play design games. Show what you made and see how others write video prompts. The community talks in Ukrainian.
- Free
- 1,504 members
- Join in 1 click
Updated · Spaceberry Foundation
Sources:
- Google Flow Help — Get started with Google Flow
- Google Flow Help — Where you can use Google Flow
- Google Flow Help — Manage your Google Flow credits
- Google Flow Help — Models & supported features
- Google Flow Help — Create videos in Google Flow
- Gemini Apps Help — Generate videos
- Runway Help — Free plan details
- Runway Help — How do credits work?
- Runway Help — Getting Started with Generative Video
- Runway Help — Creating with Gen-4 Video
- Runway Help — Gen-4 Video Prompting Guide
- Adobe Help — Generate videos using text prompts (Firefly)
- Adobe Help — Generate Video FAQ (Firefly)
- Adobe Help — Writing effective text prompts for video generation
- Adobe Help — Generative credits FAQ
- OpenAI Help Center — What to know about the Sora discontinuation