Midjourney V7 I2V is 88 credits at 720p

Midjourney V7 Image to Video is 88 credits of silent 720p (28 at 480p, the rate the card leads with). Spend that sticker when the still is already a Midjourney V7 plate and you accept no soundtrack. It is not a sketch pad, not a 1080p master, and not a native-audio hero. If you only have a paragraph, you are not on this row.

This page is the credit-and-audio pick. 88 credits is the catalog price at 720p; the card leads with 28, the 480p rate. Audio is false. Max output is 720p. Category is image-to-video. An image is required. Duration chips are empty on the published card — do not write 5s or 10s as a fact. You are buying cinematic motion on a locked V7 still, not a talking clip and not a cheap T2V pass.

88 credits is the I2V sticker

Midjourney V7 the stills row is 8 credits, text-to-image, no image required. The I2V row is 88. Eleven stills at V7 is the same sticker as one move. That ratio is the decision. If the look is not signed off, generate stills. If the look is signed off and the job is motion, pay 88.

Do not average it into a per-second rate the card does not publish. The catalog row lists a flat price per resolution (28 at 480p, 88 at 720p), one resolution ceiling, audio false, image required.

What 88 buys on this slug:

• Image-to-video from an input still, with the catalog description of cinematic motion and style consistency.
• 720p ceiling. Not 1080p. Not 4K.
• Silence. Audio false. Prompting dialogue may paint words on the plate. It will not return a VO track.

What 88 does not buy:

• A text-only generate. requires_image is true. The row will not start without a plate.
• A duration picker like Hailuo’s 6s / 10s or Seedance’s 4–15s. Inspect the job. Prompting “30 seconds” does not add a chip the row does not have.
• Native audio. If the brief talks, plan TTS or a talking video row after, or pick a different family now.

The generate door for a locked plate is the AI video generator (or the image-to-video tool on the same catalog). Pick the slug. There is no Midjourney “studio” on Versely that changes the bill.

Style consistency is the only reason this sticker is rational. V7 stills have a look. I2V is sold as keeping cinematic motion and style from that look. That only helps if the input image is the look you meant. A random product photo on this row is 88 credits of motion on someone else’s lighting. A V7 still you already approved is the case the row was listed for.

Silent 720p is the contract

Audio false means you will add sound later, or you will not. Both are legal.

Later, in practice:

• Speech: Gemini 3.1 Flash TTS at 12 credits per 1,000 characters, then captions. Lay the voice under the plate. Do not lipsync a still that has no mouth.
• Music: a music row, not a prompt clause on I2V.
• Captions: always, if the clip hits a muted feed. Burned-in captions are pixels. Native audio would not have saved you on mute anyway.

720p is the delivery ceiling. Social cuts often live here. A 4K insertion order does not. If the master must be 1080p or 4K, this is the wrong row, or it is a look-dev you will replace. Do not treat 88 credits as an upscale. It is a generate at 720p.

Compare neighbouring stickers so 88 is a choice:

• Hailuo 2.3 Fast: 12 credits (was 23), I2V, 6s / 10s, discounted, audio unset, image required. Cheaper silent motion. Not V7 style lock.
• Hailuo 2.3 Pro: 40 credits, 1080p, 6s / 10s, audio false, image required. Silent, sharper listing, still not V7.
• Kling 3 Turbo Text to Video: 22 credits (was 27), T2V, silent, 1080p, 3–15s. No still, no V7 identity.
• Runway Gen-4.5: 200 credits, T2V + I2V, 5s or 10s, audio false. Control, still silent, no 720p-only V7 contract.
• Veo 3.1: 64 credits, T2V, native audio, 4K, 4 / 6 / 8s. Talking cinematic. Not a V7 still mover.
• Flux 3 Text to Video: 34 credits, T2V, native audio, 5–20s. A sentence with a soundtrack. Not I2V.

88 sits in a narrow hole: V7 still, silent, 720p, dearer than both Hailuo I2V neighbours, cheaper than Runway’s 200, and cheap relative to wasting an 8-second Veo take on a plate that only needed a slow push.

Native audio is a catalog flag. This row does not have it. Teams that “always buy the premium video model” land here because the stills were Midjourney and the word premium felt like a soundtrack. It is not.

When the still is already V7

Use the 88-credit row when all of these are true:

1. The still was generated (or finished) on Midjourney V7 at 8 credits, or it is a plate you would accept as a V7 still: composition, character, lighting, the look you would print.
2. The still is approved. Label lettering holds. Face holds. You would not regenerate the still for a crop.
3. The destination accepts 720p, or you are look-deving motion before a different 1080p family.
4. Nobody has to speak in the generate. Captions and TTS can carry the line.
5. You want this image to move, not a cousin invented from a prompt.

Then pay 88 once. Review the motion. If the camera drifts off the label, change the motion brief, not the still. If the still was wrong, go back to 8-credit stills. Do not spend a second 88 hunting a look the I2V row cannot invent.

I2V inherits aspect from the plate more often than people plan for. A 16:9 V7 still becomes a 16:9 clip. If you will ship 9:16, crop the still first. Resizing after is a strip through a wide composition. The 88-credit file is a bad time to discover that.

Batch rule: do not run I2V on five almost-V7 stills to see which lighting you like. That is 440 credits of motion on undecided plates. Run five stills at 8. Pick one. Move it.

If the still is a product photo from a catalog shoot, this is usually the wrong family. Hailuo Fast at 12 will move a pack. You are not paying 88 for V7 consistency you never had.

If the still is V7 but the brief is a 15-second talking hero, this is also the wrong family. Length is not a chip here, and audio is false. Pick a talking T2V row and accept that the V7 still is a style reference in your head, not an input this slug will voice.

When 88 is the wrong row

Text only. Wrong category. V7 stills at 8, or a T2V row.

You need a soundtrack in the generate. Audio false. Veo 3.1, Flux 3 T2V, LTX 2.3 Text to Video Fast at 20 credits with optional native audio, 6–20s, 4K listing — those are talking (or optionally talking) cards. This is not.

You need 1080p or 4K as the master. 720p ceiling. Hailuo Pro lists 1080p at 40, still silent. Veo lists 4K at 64, with audio.

You need a 6s / 10s picker, or a 15-second hold. No duration chips. Hailuo for the staircase. Kling Turbo or Seedance-family T2V for longer silent or native holds — check the card.

The still is not V7 and you do not care about V7 style. 12-credit Fast I2V. 22-credit Turbo T2V. 88 is a style lock, not a default premium.

You already have a keeper at 720p. Do not rerun I2V to “improve” it. You will get a cousin and another 88.

Caption whatever you ship to a feed. Silent 720p on mute without type is a moving painting. Add captions to a video after picture lock. Safe area still applies. 88 credits does not buy a lower-third.

The honest sentence for a producer: “We have a signed V7 still, we will accept 720p, nobody speaks in-camera, motion is the remaining job.” If any clause is false, pick another slug. If all are true, 88 is the listed price of that remaining job.

Run the V7 still through I2V from Versely when that sentence is true. Keep the 8-credit stills row for everything before it.

FAQ

Is Midjourney V7 Image to Video 88 credits with native audio?

No. 88 credits, image-to-video, 720p, audio false, image required. Speech is a later TTS or a different video row.

Can I skip the V7 stills row and prompt video directly?

Not on this slug. Category is image-to-video. Generate or finish a still first — 8 credits on Midjourney V7 if you want that look — then spend 88 moving it.

Why not always use Hailuo Fast at 12 credits instead?

Fast is cheaper silent I2V from any still, 6s or 10s. It is not a V7 style lock and it does not publish a 720p V7 contract. Use Fast when the plate is a pack shot. Use 88 when the plate is already V7 and that look has to survive the move.