Your static pack shot won a carousel test last month. This week the same SKU needs motion hooks for Reels — and the creator invoice is still pending. That gap is exactly why teams open kreadoai image to video first: one product photo in, model-selected motion out, MP4 ready for Ads Manager in minutes.
Search results in 2026 still rank generic "best AI video" listicles or single-model API docs. Few map the KreadoAI workflow: upload a frame on Image to Video, pick Seedance / Hailuo / Veo / Sora 2, set ratio and duration, export.
What KreadoAI Image to Video Does
Image to Video is KreadoAI's multi-model motion generator. You upload a still image — pack shot, lifestyle frame, or scraped listing still — and choose a video model to animate it.
Core capabilities:
- Model picker — Seedance 2.0, Hailuo, Veo 3, Sora 2, Kling 3.0, Wan, and others in one account.
- Reference lock — uploaded frame anchors product geometry through orbit, push-in, and stamp motions.
- Ratio control — 9:16, 16:9, 1:1, 4:5 at generation time (one ratio per render).
- Duration bands — typically 4–15 seconds depending on model; Hailuo excels at 6S hooks.
- Optional native audio — SFX and off-screen VO on supported models when generateAudio is enabled.
- Same wallet as static tools — pair clips with Ad Creative Generator outputs in one brief.
Typical use cases: TikTok/Reels hooks from pack shots, product hero orbits for PDP and landing pages, motion variants for A/B tests, presentation and brand storytelling clips.
Image to Video vs URL to Video (Pick the Right Path)
KreadoAI runs two video pipelines. Mixing them in one step is the most common setup mistake.
| Image to Video (Path B) | URL to Video (Path A) | |
|---|---|---|
| Input | Product photo / still frame | Product listing URL |
| Model selection | Yes — Seedance, Hailuo, Veo, Sora, etc. | No model picker |
| Output | Motion clip from your frame | Script-to-final conversion ad (~10 min) |
| Best for | Hook volume, orbit hero, controlled motion | Listing → full ad with VO and pacing |
Rule of thumb: need to control which model animates your pack shot → Image to Video. Need a full conversion video from a Shopify or Amazon link → URL to Video.
Hybrid workflow teams use often: scrape a clean still via URL to Video, then upload that frame to Image to Video with Seedance for a 9:16 hook — URL scrape for asset, I2V for motion control.
No product photo yet? Start with Text to Video for concept tests; switch to Image to Video once pack photography lands.
The PAC Lock Method (Prompt Framework)
Most failed I2V renders trace back to vague prompts that ask the model to redesign the product. Use PAC Lock — Product anchor, Action-only motion, Constraints — on every brief.
P — Product anchor
Upload a clean, centered pack shot (1000×1000+ recommended). Prompt must reference "matches uploaded image" or "@Image1 geometry."
A — Action-only motion
Describe camera path and product movement only. Do not restyle color, packaging, or background unless intentional.
C — Constraints
Lock ratio, duration, negative space for ad overlays, and quality floor: no flicker, no morph, no on-screen text unless requested.
Example PAC Lock prompt (skincare pump, 9:16, 10S, Seedance):
10-second vertical hook. White pump bottle matches uploaded image exactly. Single pump press, subtle product glow, slow push-in — geometry unchanged. Top 15% clear for headline overlay. No text in frame. Stable motion, sharp label.
One variable per regen. Four variants in one session is normal.
Which Model to Pick (Routing Table)
Route by brief, not brand loyalty — same KreadoAI wallet for all models.
| Brief | Model | Why |
|---|---|---|
| 6S single-action TikTok Spark hook | Hailuo for Ads | Fastest 6S iteration from stills |
| 8–15S hook + reference lock + VO | Seedance 2.0 for Ads | Multi-beat arcs, generateAudio lines |
| Photoreal glass, pour, latch + SFX | Veo 3 for Ads | Premium texture, synced product sound |
| Crema, fizz, complex physics | Sora 2 for Ads | Physics choreography on hero SKUs |
| Cinematic product orbit, longer hold | Kling 3.0 | Smooth orbit on flagship frames |
For deep model-specific walkthroughs, see How to Use Seedance 2 for Ads and best AI video generators for ad creative.
How to Use Image to Video on KreadoAI: Step by Step
We timed a skincare pump → 9:16 hook in about four minutes end to end.
Step 1 — Open Image to Video and Upload Frame
- Go to kreadoai.com/image-to-video or ai/aivideogen.
- Upload pack shot — product centered, minimal clutter, 1000×1000+ resolution.
Step 2 — Select Model, Ratio, Duration, Audio
- Model: Seedance 2.0 for hook volume; Hailuo for 6S; Veo 3 for sensory hero shots.
- Ratio: 9:16 for Reels/TikTok; 16:9 for feed and YouTube; 1:1 for carousel backup.
- Duration: 6S (Hailuo) or 8–12S (Seedance/Veo) for cold traffic.
- generateAudio: enable for pump, spray, stamp SFX or off-screen VO when the model supports it.
- Paste PAC Lock prompt — action and camera only, geometry locked to upload.
Step 3 — QA Like an Ad Lead
Before trafficking, check:
- Label readable mid-motion
- Hook visible in first second
- Top 15% and bottom 20% clear on 9:16
- No morphing between product shapes
- Audio sync if generateAudio is on
One regen per variable — ratio, duration, or motion verb — not all three at once.
Step 4 — Export, Name, Pair Statics
- Download MP4 at native resolution.
- Name
sku_model_ratio_hook_v#(e.g.serum_seedance_916_pump_v2). - Pair with Ad Creative Generator statics for retargeting in the same campaign.
Export Specs (Trafficking Reference)
| Placement | Ratio | Duration sweet spot |
|---|---|---|
| TikTok In-Feed | 9:16 | 6–15 s |
| Meta Reels | 9:16 | 9–15 s |
| Meta in-stream | 16:9 | 10–15 s |
| YouTube pre-roll | 16:9 | 12–15 s |
| Instagram Feed backup | 1:1 or 4:5 | 6–12 s |
Pick ratio at generation — regenerate at 9:16 after a 16:9 winner; do not stretch in post.
FAQ
What is KreadoAI Image to Video?
It is KreadoAI's multi-model tool that animates a still image into a short video clip. You choose the motion model, ratio, and duration — unlike URL to Video, which builds a full conversion ad from a listing link without a model picker.
Is KreadoAI Image to Video free?
KreadoAI offers free-tier credits; Image to Video consumes K coins per generation. Confirm live pricing on image-to-video.
Image to Video vs URL to Video — which do I use?
Image to Video when you have a pack shot and need model-controlled motion. URL to Video when you want script-to-final conversion video from a product URL — no Seedance/Hailuo/Veo picker on that path.
Which model is best for product ads?
Hailuo for 6S hooks, Seedance for 8–15S volume with reference lock, Veo 3 for photoreal pour/glass moments, Sora 2 for complex physics demos. Most teams rotate models by brief, not one model for everything.
How long can Image to Video clips be?
Typically 4–15 seconds depending on model. Hailuo caps around 6S; Seedance and Veo support longer multi-beat arcs up to ~15S.
Can I use a product URL directly in Image to Video?
No — Image to Video requires an uploaded image. Paste the URL into URL to Video to extract stills, then upload the best frame here.
What image size works best?
1000×1000 pixels or larger, product centered, clean background. Higher resolution improves label fidelity through motion.
How is this different from Text to Video?
Text to Video generates the first frame from a prompt — useful for concept tests. Image to Video locks geometry to your upload — use it whenever a real pack shot exists.
Image to Video is the motion layer most DTC teams run daily — hooks, orbits, pours — while URL to Video handles full listing-to-ad conversion in a separate pipeline.
Open Image to Video on KreadoAI, upload one pack shot, and run PAC Lock on your first 9:16 hook before lunch.
