Veo turns words or images into video with sound in the output. Start from a prompt, from first and last frames, or from image references, output at 720p to 4K in 4, 6, or 8 seconds, and see the credit cost before you generate.
Video comes back with sound, suppressed only for sensitive scenes.
Guide the shot with first/last frames and one to three image references.
Cinematic video with sound, from text or from a reference frame. Real examples. Every clip here began as text or a single reference frame.
Veo is Google DeepMind's video model that turns text or images into short video with sound in the output. It handles cinematic motion, natural camera movement, and matching audio in a single generation, so a clip comes back closer to finished than a silent render. Beyond text-to-video, Veo accepts first and last frames plus one to three image references, which lets you fix the start, the end, or the look instead of re-rolling a prompt.
Output runs 720p, 1080p, or 4K at 4, 6, or 8 seconds, in 16:9 or 9:16. On Kavel, Veo runs across Lite and Fast tiers, the credit estimate updates before you commit, and failed generations are refunded automatically (unless it broke the content policy). Because sound arrives in the same render, a short scene that would normally need a separate audio pass and an editing suite comes back closer to finished, ready to drop into a reel or a rough cut.
Creative engine
Choose an input method, set resolution, aspect ratio, and length, and the credit estimate updates before you generate.
Text-to-video turns a written description into a moving shot with sound, matching motion to the scene you describe.
First and last frames plus one to three image references let you steer the start, end, and look of the clip.
Lite and Fast tiers trade cost and speed, so you can draft cheaply and step up only when a clip is worth it.
Reach for Veo when you want cinematic motion and sound in one generation, and when a first or last frame should anchor the shot. The Lite and Fast tiers keep drafting affordable while the higher-resolution output covers the final cut. For teams that iterate, that means a cheap first pass to lock the framing and motion, then a single higher-resolution render once the shot is right, with the credit cost visible at every step so a review never turns into a surprise bill.
Clips come back with sound, so a scene needs no separate audio pass unless the content is sensitive.
First and last frames plus up to three image references anchor the start, end, and look of the shot.
Pick a lower-cost tier to draft, then step up when a clip is worth the higher resolution.
From text-to-video to frame-guided motion with audio, the model covers a full short-clip workflow.
Turn a written scene into a moving shot with cinematic motion.
Pin a still to the first or last frame to control the motion path.
Pass one to three image references to direct the look of the clip.
Sound is included in the result, suppressed only for sensitive scenes.
Render at 720p, 1080p, or 4K in 16:9 or 9:16.
The generation parameters for this model, from the provider's published spec.
Four steps from prompt to a finished clip.
Start from text-to-video, or from first/last frames and references.
Describe the subject, action, camera, and mood of the shot.
Choose the resolution, aspect ratio, and clip length.
Confirm the estimate, run it, and download the result.
From cinematic shots to vertical social clips, here is where the model earns its place in a workflow.
Generate film-style shots with natural camera movement and sound for a scene, a trailer, or a mood piece, without booking a crew or a location.
Make 9:16 clips for reels and shorts, with audio already in the file and nothing to add in post.
Turn a key frame into a short, polished clip for a campaign or product listing, with sound already in the file and no editing suite required.
Pin a first and last frame to preview a shot before a real shoot, so a director or client can react to motion and timing instead of a static board.
One credit pool covers Nano Banana images and Seedance and Veo video. The cost shows before every run, and failed jobs are not charged. Use a subscription for ongoing work, or a one-time pack when you just need to top up.
What's included
Secure checkout by Stripe. Card details never touch our servers.
What's included
Secure checkout by Stripe. Card details never touch our servers.
What's included
Secure checkout by Stripe. Card details never touch our servers.
What's included
Secure checkout by Stripe. Card details never touch our servers.
One-time top-ups — buy extra credits any time you run low.
What's included
Secure checkout by Stripe. Card details never touch our servers.
What's included
Secure checkout by Stripe. Card details never touch our servers.
What's included
Secure checkout by Stripe. Card details never touch our servers.
Charges appear as “KAVEL AI” on your card statement. You can cancel any time from Settings → Billing; cancellation takes effect at the end of the period you already paid for.
Powered by
Use one balance across every supported image and video model — eleven of them today, listed below — and check the credit cost before each request.
Video models
Image models
Model credit guide
You're only charged for successful generations. The exact estimate in the generator varies by model, length, resolution, audio, and number of images.
| Type | Model | Credit cost |
|---|---|---|
| Video | Seedance 2.5 | The long-take tier, priced per second: 5s 480p ≈ 211 credits, 5s 720p ≈ 473. A full 30s take runs ≈ 1,260 at 480p and ≈ 2,835 at 720p. Supplying a reference clip lowers the per-second rate. |
| Video | Seedance 2.0 | 5s 720p image-to-video ≈ 188 credits; text-to-video ≈ 308 credits. Scales with resolution and length. |
| Video | Seedance 2 Fast | Faster and lower cost. 5s 720p text-to-video ≈ 248 credits. |
| Video | Seedance 2 Mini | The cheapest Seedance tier. 5s 720p ≈ 154 credits, 5s 480p ≈ 72 credits. |
| Video | Seedance 1.5 Pro | Audio doubles the rate. 5s 720p ≈ 27 credits silent, ≈ 53 with audio; 1080p ≈ 57 and ≈ 113. |
| Video | Veo 3.1 | Billed per video, not per second (Lite tier). About 45 credits at 720p and 53 at 1080p. |
| Video | Kling 3.0 | Audio raises the rate. 5s 720p ≈ 105 credits silent, ≈ 150 with audio; 1080p ≈ 135 and ≈ 203. |
| Video | MiniMax H3 | Fixed 2K, no resolution ladder. Priced per second — a 5s clip ≈ 158 credits. |
| Image | Nano Banana 2 | Generate or edit from text and images. About 8 credits per 1K image, 12 at 2K, 18 at 4K. |
| Image | Nano Banana Pro | Consistent run times across generations. About 12 credits per 1K or 2K image, 21 at 4K. |
| Image | Nano Banana 2 Lite | Faster, simpler variant. A flat 8 credits per image at every resolution. Start here to test. |
| Image | GPT Image 2 | The lowest-cost image model here. About 3 credits per 1K image, 6 at 2K, 12 at 4K. |
| Image | Seedream 5.0 Lite | Flat pricing — about 8 credits per image whether you generate or edit. |
Common questions about the Veo AI video generator.
Veo is Google DeepMind's video model that turns text or images into short video with sound. It supports first and last frames plus one to three image references. Output runs 720p–4K at 4, 6, or 8 seconds.
Yes. Sound is included in the output, so most clips come back with matching audio. Audio is suppressed only for sensitive scenes.
Credits depend on the tier and resolution. Lite at 1080p is about 34 credits, for example. The estimate shows before you generate, and failed generations are free.
Yes. Pass a first or last frame, or one to three image references, and the model generates a moving clip from them. Reference generation runs at 8 seconds.
Lite and Fast are cost-and-speed tiers of the same model. Use a lower tier to draft cheaply, then step up when a clip is worth the higher resolution.
Describe the subject, motion, camera work, and mood in detail. Pinning a first or last frame keeps the shot on track from start to finish.
Open the generator, pass text or an image, check the credits, and generate.
Pick the resolution and aspect ratio your platform needs.
