Finish
Spend the premium pass on the winner
Bring the hook, product reveal, or story beat here after the rough direction is already clear and the clip needs to look closer to final.
Reach for this model when the clip has to sell a moment: a product reveal with real camera intention, a founder ad with natural motion, a scene that carries atmosphere, sound, and continuity from the first frame to the last. Built on Google DeepMind's Veo, it follows detailed direction on lighting, movement, and audio in one pass.
Free to start · no credit card needed for a draft
Veo 3.1
Text, image, frame, and reference video generation
Preview the kind of motion you can judge before rendering: polished camera movement, believable scene continuity, native sound direction, and promptable visual detail in short video clips.
A studio orbit shot glides past a matte black wireless earbud on a reflective plinth as soft gradient light travels across the body. Clean ambient tone, a tiny mechanical click, and a faint rush of camera movement sell the premium commercial feel.
Veo 3.1 is the higher-cost lane for final-looking clips. Reach for it when a shot needs text to video, image to video, first/last-frame control, and reference support in one polished pass — with director-level camera language and grounded physics running underneath it all.
Finish
Bring the hook, product reveal, or story beat here after the rough direction is already clear and the clip needs to look closer to final.
Reference
Add an image reference when the product, character, setting, or brand frame should guide the output instead of relying on text alone.
Frames
Pin the opening image and the final landing frame when the shot needs to start and end exactly where the storyboard says.
Prompt
Name the subject, camera move, action, light, and sound cue. The model works best when the prompt reads like one clear shot, not a list of mood words.
1080p
Cinematic-grade output ready for client review
~8s
Short-form length, extendable in editing
Native
Dialogue, sound effects and ambience in one pass
4
Text, image, first/last frame, reference
Use these as long-form briefs: each one names the opening frame, lens behavior, scene beat, finish, and audio cue, so the model has a complete shot to direct rather than a mood board.
Prompt 01
A cinematic 16:9 product film of a brushed steel watch resting on dark green leather. Start with an extreme macro of the second hand moving across the dial, then dolly backward as a warm key light reveals the case edge and strap texture. Add a quiet mechanical tick, soft studio room tone, premium editorial finish, accurate metal reflections, no text or logos added.
Prompt 02
A founder stands in a quiet elevator lobby before a launch event, holding a small prototype box. Begin on her hands adjusting the box, tilt up to her face as she delivers one calm spoken line, then end on the doors opening behind her. Natural lobby reverb, subtle fabric movement, confident startup film tone, stable handheld camera, keep identity and outfit consistent.
Prompt 03
A black car stops outside a boutique hotel at night while rain shines on the pavement. The camera starts low near the tire reflection, tracks with a polished suitcase rolling past, then racks focus to the hotel entrance glowing in warm light. Include rain ambience, distant traffic, restrained cinematic music cue, elegant travel commercial mood, no signage text.
Prompt 04
A slow interior walkthrough of a minimalist kitchen at sunrise. Start with light sliding across stone countertops, glide past a chrome faucet, then reveal a wide window with city rooftops beyond. Keep vertical lines straight, materials crisp, reflections natural, add quiet room tone and a soft appliance hum, high-end real estate video finish.
Prompt 05
A translucent perfume bottle stands on pale marble with a silk ribbon moving in a light breeze. The camera begins behind drifting mist, pushes through it, then lands on the bottle silhouette as golden side light catches the glass. Add a delicate spray sound, soft cloth rustle, luxury beauty ad pacing, clean background, no text on screen.
Prompt 06
A barista prepares an espresso in a dim morning cafe. Start on coffee grounds falling into the portafilter, cutless camera move follows the tamp, then pushes close as espresso streams into a small ceramic cup. Include grinder hum, cup contact, low cafe ambience, rich contrast, shallow depth of field, finished brand-film look.
Start with the subject and the first frame, then describe how the camera moves, what changes, and what the last frame should feel like. The model responds better when the clip has a small arc.
Ambient sound, product texture, dialogue, or a music cue can make the clip easier to judge. Keep it grounded. A few precise sound notes beat asking for a full soundtrack.
Keep the same subject and camera move, then test light, pacing, background, or product angle. This keeps your tests comparable instead of turning every render into a new idea.
Same shot brief, four different jobs. Resolution and clip-length figures are model targets to confirm before launch.
| Feature | Veo 3.1 | Veo 3.1 Lite | Seedance 2 | Seedance 2 Fast |
|---|---|---|---|---|
| Best for | Premium cinematic clips | Fast Veo-style drafts | Reference-driven motion | High-volume variations |
| Finish | Highest finish | Lighter pass | Flexible | Quick draft |
| Audio | Native sync audio | Sync audio | Optional audio | Simple audio |
| Speed | Standard | Faster | Standard | Fastest |
| Resolution | Up to 1080p | Up to 1080p | Up to 1080p | 480p–720p |
| Top use | Hero ads, brand films | Social hooks, iteration | Ecommerce, character | Batch ad testing |
Veo 3.1 Lite
The Lite tier is the practical generator for fast social ads, image to video drafts, product motion tests, and creator clips. Reach for it when you want the same camera logic and audio-aware direction as the full model, but the real job right now is finding the winning hook before anyone over-polishes the wrong idea.
Compare this modelSeedance 2
Seedance 2 reads text, images, video clips, and audio in a single prompt, then keeps your product, character, motion style, and sound locked to the references you gave it. Use text to video to start from an idea, image to video to animate a still, or multi-reference generation when a shot needs several anchors at once.
Compare this model