Kling 3.0

Kling 3.0, officially Kling VIDEO 3.0, is Kuaishou's AI video model, launched in early February 2026. It makes 3–15 second clips from a text prompt or a start image, with optional native sound and lip-synced dialogue. On API platforms it appears as kwaivgi/kling-v3.0 in Standard, Pro and 4K versions; Kling also offers a separate VIDEO 3.0 Omni model.

All three take the same inputs and make 3–15 second clips in 16:9, 9:16 or 1:1. Standard is the cheapest (25.2 credits a second silent on Kenerate), Pro is the higher-quality default (33.6), and 4K renders at 4K resolution for 126 credits a second with sound included. Draft on Standard, then re-run the prompt you like on Pro or 4K.

It's billed per second. Standard is 25.2 credits a second silent or 37.8 with sound, Pro is 33.6 or 50.4, and 4K is 126 with sound included. A 5-second clip costs 126 credits on silent Standard, 252 on Pro with sound and 630 in 4K. Credits come in one-time packs that never expire, starting at $15 for 1,300 credits.

Not on Kenerate's free plan. Video generation needs a paid plan, which means buying any one-time credit pack — there is no subscription. New accounts get 25 starter credits for images. Kling's own app has its own free allowance and paid tiers; check kling.ai for its current terms.

3 to 15 seconds per clip, in whole seconds, on all three tiers. That's up from 10 seconds on Kling 2.6. For longer pieces, generate several clips — using the last frame of one as the start frame of the next keeps the look consistent — and join them in an editor.

Yes, if you switch it on. Kling's guide says Kling 3.0 generates native audio with dialogue in Chinese, English, Japanese, Korean and Spanish, plus effects and ambience. On Kenerate sound is a toggle: it raises Standard to 37.8 and Pro to 50.4 credits a second, and the 4K tier includes it in the 126-credit rate.

Yes. Upload a start image in /app/video and Kling 3.0 animates it; add an end image as well and it moves from the first frame to the last. Image to video is the most reliable way to keep a product, a character or a face exactly as designed — make the still first, then animate it.

It's the API name for Kling 3.0 Pro used by model platforms such as WaveSpeed — kwaivgi is Kuaishou's Kling namespace, with -std and -4k for the other tiers and separate text-to-video and image-to-video endpoints. On Kenerate you don't need an API key: the same model runs in the browser as Kling v3 Pro.

Yes, as a separate mode. Kling 3.0 Motion and Kling 3.0 Pro Motion copy the movement of a 3–30 second reference video onto your character image and can keep the clip's original sound. On Kenerate they cost 37.8 and 50.4 credits a second — see the Kling motion control page.

Clips go from a 5 or 10 second choice to anything from 3 to 15 seconds, native audio gets multi-language dialogue and lip-sync, and there is now a 4K tier. Kling's guide also adds multi-shot storyboards and element references in its own app. Kling 2.6 Pro stays cheaper per second on Kenerate (21 silent, 42 with sound).