AI image to video tools take one still photo and turn it into a short moving clip. You add a prompt, pick a camera move, and the model fills in the motion. However, free limits, prices and safety rules differ a lot between apps. So this guide compares six real tools, shows the steps, and covers what you must not do.
How AI image to video works
An AI image to video tool treats your photo as the first frame of a scene. Then a video model predicts what the next frames should look like, based on your prompt. Because it learned from huge amounts of video, it can guess how water flows or a camera pans.
Your prompt guides the motion, mood and sometimes sound. In short, the photo sets the look, and the prompt sets the action.
Best AI image to video tools right now
We checked each official pricing page on 30 October 2026. Prices are in US dollars, and plans change often, so check again before you pay.
Google Veo in Gemini and Flow
Google added photo to video in Gemini on 10 July 2025, powered by Veo 3. At launch, it made eight second clips with sound. Today, Google’s plans page lists video generation only on paid plans: Google AI Plus ($4.99 a month), AI Pro ($19.99) or AI Ultra (from $99.99). Ultra lists Veo 3.1, while Pro lists a limited Veo 3.1 Lite trial. Flow, Google’s AI film studio, also gives credits on each paid plan.
Runway
Runway’s pricing page gives free users 125 credits once, and they do not expire. Its Gen 4.5 model costs 60 credits per five second clip, so the free pool covers about two clips. Next, the Standard plan costs $15 a month, or $12 a month billed yearly, with 625 credits and no watermarks.
Kling AI
Kuaishou launched Kling 3.0 on 6 February 2026, with image to video, clips up to 15 seconds and native audio in many languages. Then Kling 4.0 arrived on 15 October 2026, with clips up to 30 seconds. Kling’s credit policy sets the standard rate at $1 for 66 credits. Still, watermark removal is a premium feature, so free clips may carry a mark.
Luma AI
By comparison, Luma’s pricing page no longer lists a free plan. Its Plus plan costs $30 a month, or $25 a month billed yearly, with 10,000 credits and commercial use. In addition, it offers its own Ray3.2 and Ray3.14 models plus partner models like Kling and Veo.
Pika
Pika keeps a free plan, but it includes no monthly credits. Instead, free users buy credit packs, and commercial use is not allowed. Meanwhile, the Starter plan costs $10 a month, or $8 billed yearly, and gives 900 credits.
Adobe Firefly
Likewise, Firefly gives free daily generations across image, video and audio. The Standard plan costs $9.99 a month for 2,000 credits, or up to 20 five second videos. Also, it includes partner video models from Runway, Kling and Luma.
What happened to OpenAI Sora?
Notably, OpenAI has shut down Sora as a consumer product. Its own Sora 2 page now says the product is no longer available as of 26 April 2026. According to The Decoder, OpenAI set the Sora API to close on 24 September 2026. So we left it out of this list.
AI image to video tools compared
Here is a quick view. Note that “free” may mean one time credits, daily credits, or no credits at all.
| Tool | Free option | Starting paid price | Clip notes |
|---|---|---|---|
| Gemini / Flow (Veo) | No video on free Gemini | $4.99/month (AI Plus) | Visible watermark plus SynthID |
| Runway | 125 one time credits | $12/month (yearly) | Gen 4.5 at 60 credits per 5s |
| Kling AI | Check app offers | $1 = 66 credits | Up to 30s in one pass (4.0) |
| Luma AI | No free plan listed | $25/month (yearly) | Ray3.2, plus partner models |
| Pika | Free plan, no monthly credits | $8/month (yearly) | No commercial use on free |
| Adobe Firefly | Free daily generations | $9.99/month | Content Credentials attached |
How to animate a photo step by step
- Pick a strong photo. First, choose a sharp image with one clear subject.
- Use your own image. Also, only upload photos you took or have the rights to use.
- Choose the format. Next, pick 9:16 for Reels and Shorts or 16:9 for YouTube.
- Write a short prompt. Then describe one action and one camera move, such as “slow push in, leaves sway”.
- Generate two or three versions. After that, compare them, because each run gives different motion.
- Download and edit. Finally, trim the best clip and add music or text.
For short social clips, see our guide to the best AI reel maker apps. If you post on YouTube, our AI for YouTube guide covers scripts, thumbnails and editing.
Free: AI Tools Cheat Sheet 2026
24 free tools sorted by task, a simple prompt formula and 10 prompts you can copy. We send the PDF straight to your inbox.
Sent! Check your inbox
We emailed the AI Tools Cheat Sheet to you. Not there in 2 minutes? Look in Spam or Promotions.
Prompt tips for camera motion
Above all, clear camera words give the model a plan.
- Push in or pull out. For example, “slow dolly in toward the face” adds drama.
- Pan or tilt. Say “pan left across the skyline” for wide scenes.
- Orbit. Also, “camera orbits the product” works well for objects.
- Static shot. Sometimes “locked camera, only the water moves” looks the most real.
In general, keep it to one main move per clip. Because clips are short, too many actions make the motion messy.
Rights and safety for AI image to video
Never animate a photo of a real person without their clear consent. A moving face can mislead viewers or hurt someone, and it may break laws in some places. Also, avoid photos of children and strangers.
Next, respect copyright and use only images you own or license. Also, Pika’s free plan does not allow commercial use, so check terms before client work.
Finally, keep AI labels in place. Google marks Gemini video clips with a visible watermark and an invisible SynthID mark. Adobe attaches Content Credentials, which follow the C2PA open standard for showing where content came from. So do not crop or strip these marks, and tell your audience when a clip is AI made.
Bottom line
Start with Runway’s free credits or Firefly’s daily generations to learn how prompts change motion. Then, if you need longer AI image to video clips, compare Kling 4.0 and Google’s Veo plans. In short, pick one tool, learn its camera words, and always use photos you have the right to animate.
Also read: AI 3D Model Generator: Best Tools to Turn Text and Photos Into 3D
Frequently asked questions
Is AI image to video free?
Some tools offer free use with limits. Runway gives 125 one time credits, and Adobe Firefly gives free daily generations. However, video in the Gemini app needs a paid Google AI plan.
How long can an AI video from a photo be?
It depends on the model. Gemini’s photo to video feature launched with eight second clips, while Kling 4.0 makes up to 30 seconds.
Can I animate a photo of another person?
Only with their clear consent. Animating someone without permission can mislead viewers and may break the law where you live. Also, read the tool’s own rules on real people before you upload.
Is OpenAI Sora still available?
No. OpenAI says the Sora product is no longer available as of 26 April 2026. Runway, Kling, Veo and Firefly are good alternatives.
