How to Use HappyHorse Image-to-Video: Specs, Credit Costs and Prompts

HappyHorse Image-to-Video turns exactly one still image into a 4 to 15 second clip at 720p or 1080p, priced from 46 credits (about $2.30) for a 4-second 720p render up to 295 credits (about $14.75) for a 15-second 1080p render. It is the HappyHorse family’s single-image animation mode: you upload one picture, describe the motion you want, pick a length and a resolution, and the model produces the movement. There is no aspect-ratio control on this mode, because the output follows the framing of the image you hand it.

What HappyHorse Image-to-Video is best at

The model earns its place on two specifics. The first is the duration ceiling: 15 seconds, selectable at every whole second from 4. Most image-animation models cap out at 5 or 10 seconds, which forces you to either accept a rushed camera move or stitch clips together. With 15 seconds available in one-second steps, a slow push-in, a drifting parallax shot, or a two-beat action can play out at its natural pace and be cut to fit an edit exactly.

The second is that it takes any still. A phone photo, a scanned print, a product shot, a 3D render, or an image you generated minutes earlier in AI image generation all work the same way. That makes it the natural back half of a two-step pipeline: make the frame you want with an image model where iteration is cheap and fast, then spend video credits only once, on a frame you already know is right.

HappyHorse carries a 4.8 rating in the VdoBloom catalog and sits in the MODERATE content tier — standard filtering, not the strictest tier and not the most flexible. Portraits, products, landscapes, fashion and fitness content generate without friction. If you are animating a photo of a real person, you must have that person’s consent before you generate or publish the clip.

The real specs

Setting	Image-to-Video 1.0	Image-to-Video 1.1
Input images	Exactly 1	Exactly 1
Prompt	Required — describes the motion	Required — describes the motion
Durations	4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 seconds	Identical — 4 to 15 seconds, one-second steps
Resolutions	720p, 1080p	720p, 1080p
Aspect ratio control	None — follows the source image	None — follows the source image
Content tier	MODERATE	MODERATE
Catalog rating	4.8	4.8
Credit cost	See the credit table below	Identical to 1.0

That last row matters more than it looks. Because 1.0 and 1.1 are priced the same and share the same duration and resolution range, choosing between them is purely a question of output character, not budget or capability. Run the same source image and the same prompt through both at 4 seconds and 720p — 46 credits each — and pick the one that reads better for your subject.

The aspect-ratio row surprises people. The text-to-video and reference-to-video modes in the same family expose an aspect ratio picker; image-to-video does not, on either version. The framing you get is the framing you upload, so crop your source before uploading.

Exact credit cost per tier

Credits convert at a flat rate on the Lite plan: $15 for 300 credits means one credit is $0.05. Every figure below is the real cost charged before generation, with the dollar equivalent at that rate.

Duration	720p credits	720p cost	1080p credits	1080p cost
4 seconds	46	$2.30	79	$3.95
5 seconds	58	$2.90	99	$4.95
6 seconds	69	$3.45	119	$5.95
8 seconds	92	$4.60	158	$7.90
10 seconds	115	$5.75	197	$9.85
12 seconds	138	$6.90	236	$11.80
15 seconds	173	$8.65	295	$14.75

Two things fall out of these numbers. First, 1080p costs roughly 70 percent more than 720p at every length, so the honest workflow is to test at 720p and only re-run the keeper at 1080p. Second, HappyHorse is a premium-tier model: the 10 free credits every new account receives will not cover a single run, so plan on a credit pack or a subscription if this is the model you want. Full tiers are on the pricing page, where one-time packs start at $2.49 for 75 credits and never expire.

How to prompt it well

Image-to-video prompting is not image prompting. The picture already establishes the subject, the lighting, the wardrobe and the composition — repeating all of that in the prompt wastes the model’s attention and often causes it to redraw details it should have left alone. Write about change instead of appearance.

• Name one camera move. Slow push-in, gentle pull-back, lateral dolly left, subtle handheld drift. One move, not three.
• Name one subject action. She turns her head toward camera and smiles. The steam rises from the cup. Hair lifts in the breeze.
• Name the ambient motion. Fabric sways, leaves move, water ripples, dust catches the light. This is what separates a living clip from a slow zoom on a photo.
• Set the pace explicitly. Words like slow, gradual, subtle and steady genuinely reduce the jitter that shows up when a model tries to fill a long duration with too much event.
• Say what stays fixed. Identity and expression remain unchanged. Background stays static. Instructions about what should not move are cheap insurance against drift on longer clips.

The most common failure at 12 to 15 seconds is over-instruction: a prompt with four actions in it forces the model to cram, and the result warps. If you want a lot to happen, generate two clips of 6 to 8 seconds and cut them, which is also close to the same money — two 8-second 720p runs cost 184 credits against 173 for a single 15-second run, and give you an edit point.

A worked example

Say you have a studio photo of a jacket on a model, shot in portrait, and you need a 9:16 clip for a feed placement.

1. Crop the source to 9:16 before uploading. The mode has no aspect ratio setting, so the crop is your framing decision.
2. Open image-to-video and select HappyHorse Image-to-Video 1.0 or 1.1.
3. Upload the single image. The mode rejects multi-image uploads — it accepts exactly one.
4. Prompt: slow push-in, the model turns her shoulder toward camera, jacket fabric shifts and catches the light, subtle breathing motion, background stays static, identity unchanged.
5. Set 6 seconds at 720p. Cost: 69 credits, $3.45.
6. If the take is right, re-run the identical prompt at 1080p for 119 credits, $5.95. Total spend on a finished vertical clip: 188 credits, about $9.40.

When to pick a different model

HappyHorse Image-to-Video is the wrong tool in three situations, and being honest about them saves credits.

You have no source image. Use HappyHorse Text-to-Video instead. It shares the same 4 to 15 second range and the same credit table, and it exposes an aspect ratio picker — 16:9, 9:16, 1:1, 4:3 and 3:4 on 1.0, with 4:5, 5:4, 21:9 and 9:21 added on 1.1.

You need the same character across several clips. Image-to-video animates your exact frame and nothing more. HappyHorse Reference-to-Video takes 1 to 9 reference images of a subject and generates new prompted scenes that keep that subject recognizable, which is what an episodic series actually needs.

You want a specific, repeatable motion. If the goal is a known effect — a hair flip, a catwalk turn, an outfit reveal — the dedicated effect tabs under video creation apply a tuned template instead of asking you to describe the move in prose, and they are usually cheaper per clip. Use HappyHorse when the motion is bespoke and you want the full 15-second range.

Frequently asked questions

How many images can I upload to HappyHorse Image-to-Video?

Exactly one. The endpoint validates the upload count and refuses anything other than a single image. If you want several reference images to drive one clip, that is the reference-to-video mode, which accepts 1 to 9.

Can I set the aspect ratio?

No. Neither version 1.0 nor 1.1 exposes an aspect ratio control in image-to-video — the output follows the source image. Crop your still to the ratio you want before uploading.

What is the cheapest way to test this model?

A 4-second 720p run at 46 credits, about $2.30. That is enough to judge motion quality and prompt adherence before you commit to a longer or higher-resolution render.

Is 1.1 better than 1.0?

Not on paper. Both offer 4 to 15 second durations, 720p and 1080p output, single-image input, no aspect ratio control, and identical credit pricing. The difference is output character, so test both on the same image at the 46-credit tier.

Do paid plans remove the watermark?

Yes. Paid plans download watermark-free with commercial rights, and annual billing is cheaper — Lite works out to $10 per month, and Pro to $49 per month at the 2,000-credit tier.

Can I animate a photo of someone else?

Only with that person’s consent. The MODERATE content tier blocks prohibited categories at generation time, but permission for a real person’s likeness is your responsibility, not the filter’s.