Verified 720p text-to-video
The published configuration returned H.264 around 1280×720. Duration and frame stay fixed so prompts can be compared on the same footing.
Model
Kling 4.0 Video Generation
Verified on Cyanvid: 5 seconds, 16:9, 720p, text-to-video only, AAC audio, 39 credits per clip. This page describes one published configuration.
Cyanvid publishes one verified Kling 4.0 text-to-video configuration: 5 seconds, 16:9, 720p (1280×720), about 24 fps, an AAC audio track and MP4, at 39 credits per clip. The service uses a Replicate-hosted Kling 4.0 configuration and is independent from the model authors. Thirty-second clips, 1080p, 9:16, 1:1, image-to-video, reference inputs and video editing remain unpublished until separately verified.
- Model
- Kling 4.0 · kwaivgi/kling-v4.0
- Mode
- Text-to-video only (image/reference inputs unpublished)
- Clip / audio
- 5s · 16:9 · 720p · 1280×720 · about 24 fps · AAC observed
- Credits / boundary
- 39 per clip; this one t2v variant only, other modes unpublished/unverified
Concept imagery for inspiration. Generation results vary.
Verified configuration
39 credits, shown before submit
The studio shows a 39-credit quote before submission and refunds credits when a job fails. The price follows the current Pro credit value and the measured provider cost.
Audio behavior is documented
The verified output included an AAC audio track. Check the result in your edit before adding voiceover, music or captions.
From idea to a clip
Confirm text-to-video landscape is enough
If you need an image or video input, or a vertical frame, choose another published model. Kling 4.0 is text-to-video 16:9 here.
Select Kling 4.0 in the studio
Choose the 5-second / 16:9 / 720p configuration and confirm the 39-credit quote before submitting.
Write one shot and inspect it
Use a subject, one action, one camera move and light or style. Review subject continuity, edges and the AAC track before downloading into your edit.
Advertised capabilities vs the configuration verified here
The Kling 4.0 model page lists more durations, resolutions, ratios and reference inputs. Cyanvid publishes only the one real t2v run verified here: 5s · 720p · 16:9. Other modes are marked unpublished/unverified; the 39-credit quote does not apply to them.
A few useful answers
Which Kling 4.0 configuration is published here?
One text-to-video configuration only: 5 seconds, 16:9, 720p, at 39 credits per clip. Other modes are unpublished.
Does this Kling 4.0 output have audio?
The verified MP4 included an AAC audio track. Recheck the actual file in your editing workflow before publishing.
Can I use image, reference image or reference video inputs?
No. Image-to-video, reference inputs and video editing are marked unpublished/unverified here.
Can I choose 1080p, 9:16 or a longer duration?
No. Only 5 seconds, 16:9 and 720p are listed. Other durations, resolutions and ratios need separate verification.
Why 39 credits?
The public provider estimate is $0.157 per second × 5 seconds = $0.785. Using the Pro plan rate of $29.90/400 credits (about $0.07475 per credit) and the roughly 72.85% median gross margin of current video models gives about 38.7 credits, rounded up to 39.
Kling 4.0 or the existing Kling 2.5 page?
Kling 2.5 is the verified 5-second, 16:9, 1080p, 20-credit, silent configuration. Kling 4.0 is 5 seconds, 16:9, 720p, 39 credits, with an AAC track observed. Choose by resolution, audio and budget.