Kling V2.1: the still is the contract (5s, 10s, 140cr)

Kling V2.1 is image-to-video. If the label is wrong, every second is waste.

Kling, category image-to-video, requires_image: true, durations 5s and 10s, 140 credits on the listed first-5s step (the catalog also marks a 50% discounted step at 70). Description: "Kling version 2.1 for advanced video generation." audio is unset. No max-resolution field. No aspect-ratio enum. Styles listed: Cinematic, Documentary, Abstract. This is not a text-to-video sibling with an optional picture. The still is required. If that still is the wrong product, 5s of motion is a wrong product moving, and 10s is twice the waste at the extra-second rate.

5s or 10s, from a still you would pause on

There is no 6s, no 8s, no 15s on this record. You pick 5s or 10s. The billing shape is a base for the first 5s at 140 credits, then extra seconds on the matrix. That is why a 10-second run is not "the same button." It is more time for the still to hold or to drift.

The prompt describes motion relative to the picture: camera, subject action, pacing. Re-describing the subject fights the upload. Kling's family register names camera movement and motion direction. Use that. Do not write a new character into a still of a different character and hope V2.1 negotiates.

Image to video is the door. Turn a photo into a video is the agent phrasing. Kling's roster is large; V2.1 is one older i2v row on it, not the whole brand. Best Kling model ranks the family. Best image-to-video model is the capability shelf.

140 credits is the motion bill, not a generate-from-nothing bill

140 credits for the first 5 seconds is the catalog figure. Treat it as expensive enough to inspect the still first. A 4K hero you already approved is the right input. A screenshot from a missed text-to-video take is usually the wrong one — you are paying V2.1 to animate a frame nobody signed.

Because audio is unset, do not plan on native dialogue. Because no resolution is listed, do not promise a client 4K from this SKU. Because no aspect enum is listed, crop the still to the ratio you need before you spend the 140. Asking for "widescreen cinematic" in prose will not create a 16:9 parameter this record does not have.

Kling 2.5 Turbo is a different row. Do not name-swap them. V2.1's contract on this page is image required, 5s or 10s, 140 credits.

If you wanted text-to-video, stop

Wrong label, every second waste:

• No still, just a sentence → a text-to-video row. This one will not run correctly without the image.
• A driving video of a dance → motion-control, not V2.1.
• A 20-second single take → this enum stops at 10s.
• A talking presenter with a VO → audio is unset here. Use a lipsync or native-audio generate.

The image-to-video glossary is the mechanic: frame one is the file you sent. Kling V2.1 is that mechanic at 140 credits. If frame one is not a keeper, do not buy 5 or 10 seconds of it.

FAQ

Does Kling V2.1 work from text alone?

No. requires_image is true. Category is image-to-video. The still is the contract. Text describes motion on top of it.

Why is this 140 credits for 5 seconds?

That is the catalog's first-5s figure on this row (140, discounted step 70). Extra seconds are a separate line on the matrix. A 10-second clip is a deliberate second step, not a free upgrade. Inspect the 5-second still-to-motion before you buy the other five.

Does V2.1 include native audio?

The record does not set audio: true. Plan picture. Mix later. 140 credits is not a soundtrack.

How is this different from Kling V3 motion control?

V2.1 animates a required still for 5s or 10s. Motion-control rows take a driving video and map that performance onto a still. If you have choreography on tape, you are not on V2.1. If you have a hero frame and a camera move in mind, you are.