Hailuo 2.3 Image-to-Video Standard is MiniMax’s volume tier for animating a still image, and on VdoBloom it costs 15 credits for a 6-second 768P clip, 25 credits for 10 seconds at 768P, and 30 credits for 6 seconds at 1080P. It takes one source image plus an optional motion prompt and returns a 6 or 10 second video. There is no text-only mode and no aspect ratio picker — the framing of your uploaded image is the framing of the clip.

What the Standard tier is actually for

Every image-animation workflow has two phases: finding out what motion the image can support, and rendering the version people will see. Hailuo 2.3 Standard is built for the first phase. It offers the identical duration and resolution envelope as its Pro sibling — 6 or 10 seconds, 768P or 1080P — at roughly half to two-thirds the credit cost, which means you can run the same still through five or six different motion directions for less than the price of two Pro renders.

That is the honest positioning. Standard is not a crippled model; it is the same 2.3 generation with a lighter rendering treatment. For social-sized delivery, A/B ad variants, and anything that will be watched on a phone at 9:16, plenty of creators never leave it. See the full spec sheet on the Hailuo 2.3 Image-to-Video Standard model page.

Real specs and exact credit costs

Setting	Options	Credits	Cost at $0.05/credit
6 seconds, 768P	Available	15	$0.75
10 seconds, 768P	Available	25	$1.25
6 seconds, 1080P	Available	30	$1.50
10 seconds, 1080P	Not supported by the model	—	—
Input	One image, required	—	—
Prompt	Optional motion direction	—	—
Aspect ratio	Inherited from the source image	—	—
Content filter tier	Strict	—	—

The dollar column uses VdoBloom’s Lite plan rate: $15 a month buys 300 credits, so one credit is five cents. Annual billing brings Lite to $10 a month, which pushes the effective cost of a 6-second 768P clip to about fifty cents. Full plan details are on the pricing page, and every new account starts with 10 free credits — enough to test this model twice at 768P with no card.

The one gap worth memorising: 10 seconds at 1080P does not exist on either 2.3 tier. If you need ten seconds, you are shooting 768P. If you need 1080P, you are shooting six seconds. Plan the edit around that before you write the prompt.

Choosing the right source image

Hailuo 2.3 animates what it is given; it does not invent a better composition. In practice the images that animate cleanly share a few traits:

One clear subject with breathing room. A figure filling 40–70 percent of the frame gives the model somewhere to move them.
Sharp focus on the part that should move. A soft, blurry face will smear the moment it turns.
Unambiguous depth cues. A visible floor, horizon, or background layer tells the model which way is back, which is what keeps a push-in from warping.
No text or fine logos in the moving area. Lettering is the first thing to dissolve in any generative video model.

If you are animating a photograph of a real person, you must have that person’s consent before generating or publishing the clip — that is a condition of use, not a suggestion.

How to prompt it well

Because the image already carries the entire look, your prompt should carry only the motion. Wasting words re-describing the subject is the most common mistake, and it actively hurts: a description that contradicts the image pulls the render toward an average of the two.

A prompt that works reads like a single shot instruction: she turns her head slowly toward camera and smiles, hair shifting with the movement, camera pushes in gently. Three components, in this order — subject action, secondary detail, camera move — is the pattern that holds up across attempts.

Other things that consistently help:

One action per clip. Six seconds does not fit a turn, a walk, and a gesture. Ask for one and cut between clips instead.
Name the speed. Slowly, gently, gradually. Unqualified verbs tend to render faster and jerkier than you expect.
State the camera or state that it holds. Locked-off camera is a valid instruction and produces a steadier result than saying nothing.
Avoid negatives. Do not write no distortion; write what should happen instead.

A worked example

Say you have a product shot: a perfume bottle on a stone surface, side-lit, shot in portrait. The goal is a 9:16 clip for a paid social test.

Open Image to Video, select Hailuo 2.3 Image-to-Video Standard, and upload the bottle shot. First pass: 6 seconds, 768P, 15 credits, prompt camera orbits slowly around the bottle, light sweeping across the glass, everything else still. Run it three times with three different camera moves — orbit, slow push-in, slow rise. That is 45 credits, about $2.25, to learn which move the image actually supports.

Whichever wins, re-render it at 6 seconds 1080P for 30 credits. Total spend to a finished ad variant: 75 credits, $3.75. Paid plans download it watermark-free with commercial rights.

When to pick a different model

Hailuo 2.3 Standard is the wrong tool in three situations.

When the clip is the final deliverable and the budget allows. Move to Hailuo 2.3 Image-to-Video Pro — 25 credits at 768P/6s, 45 at 768P/10s, 45 at 1080P/6s. Roughly 50–80 percent more per clip for the line’s premium rendering, which is worth it on a hero shot and wasteful on a motion test.

When you have no starting image, or need physics. Hailuo 02 does both text-to-video and image-to-video in the same 6/10 second, 768P/1080P envelope, and is tuned for weight, momentum, and collisions. It also carries the family’s more permissive content tier, where 2.3 runs strict filtering on both the image and the prompt.

When you need more than ten seconds or a specific aspect ratio. Neither 2.3 tier offers an aspect ratio control or a duration past ten seconds, so a 15-second 21:9 piece needs a different family entirely.

For how the whole Hailuo line stacks up against MiniMax’s own platform, see VdoBloom vs Hailuo AI.

Frequently asked questions

How much does one Hailuo 2.3 Standard clip cost?

15 credits for 6 seconds at 768P, 25 credits for 10 seconds at 768P, and 30 credits for 6 seconds at 1080P. At the Lite plan rate of five cents per credit that is $0.75, $1.25, and $1.50 respectively.

Can I generate 10 seconds at 1080P?

No. The model supports 10 seconds only at 768P, and 1080P only at 6 seconds. This applies to the Pro tier too.

Does Hailuo 2.3 Standard do text-to-video?

No, it is image-to-video only — a source image is required. For text prompts alone, use Hailuo 02, which handles both modes.

Can I choose the aspect ratio?

Not directly. The model has no aspect ratio setting, so the clip comes out in whatever shape your source image is. Crop the image before uploading to control the framing.

Is Standard lower resolution than Pro?

No. Both tiers offer the same 768P and 1080P options at the same durations. The difference is rendering treatment and price, not the spec sheet.

How to Use Hailuo 2.3 Image-to-Video Standard on VdoBloom