Video model · Kling AI
Kling 3.0 is the video model from Kling AI (Kuaishou), launched in February 2026 and built to tell a scene across several shots. On Deepnia you direct up to 5 shots in one 3–15 second clip, keep up to 3 characters or products identical from shot to shot, and turn on “Generated sound” to add voices, sound effects and ambience.
The video tool opens with Kling 3.0 already selected. The exact cost shows before every creation; you sign up when you generate.
Split your scene into up to 5 shots, each with its own prompt (500 characters) and a length of 1 to 12 seconds, 15 seconds in total. Kling built version 3.0 to follow these sequences, from shot-reverse-shot to cross-cutting.
Create up to 3 persistent elements from 2–4 images or one video, then call them in the prompt with @element_1, @element_2, @element_3. A mascot, a product or a character keeps the same look across the whole video.
Turn on “Generated sound” to get voices, dialogue, sound effects and ambience in the same generation.
In single-shot mode, set the start frame and, if you like, the end frame: Kling 3.0 animates the move from one to the other. Useful for a before/after or a product reveal.
Kling highlights how 3.0 preserves text from your images: signs, labels and logos stay readable, which matters for product videos.
The “Try Kling 3.0” button opens Deepnia’s video tool with the model already selected.
Turn on “Multi-shot” to open the shot editor, then write a prompt and a length for each shot.
Add a start frame (and an end frame in single-shot mode) or your persistent elements, called in the prompt with @element_1.
Choose the aspect ratio, the length and the Standard or Pro mode, turn on “Generated sound” to add voices and sound effects, check the cost shown, then create.
Copy a prompt, open the tool and adapt it to your project.
Shot 1 (4 s): wide shot of a busy food truck at dusk, string lights overhead, customers lining up. Shot 2 (5 s): close-up of loaded tacos served hot, steam rising. Shot 3 (4 s): friends laugh as they share the food, the camera slowly pulls back. Sound: street ambience, upbeat instrumental music, laughter.
@element_1 walks into a bright phone shop, hops onto the counter and proudly presents a new smartphone to the camera, then winks. Bright 3D animation style, gentle orbiting camera. Sound: cheerful sound effects, short upbeat jingle.
Smooth transition from the start frame (client with natural hair in the salon) to the end frame (the same client with long beaded braids). The camera slowly circles her, warm salon lighting. Sound: salon ambience, soft music.
@element_1 walks along a Lagos rooftop at sunset wearing @element_2, the fabric moving in the wind; lateral tracking shot, then a close-up on the embroidery. Sound: city wind, light percussion.
The perfume bottle from the start frame slowly rotates on a marble pedestal, orange blossom petals fall around it, golden reflections on the glass, the label perfectly readable. Cream background, studio lighting.
In a bright coworking space, a young founder turns to her colleague and says: "We launch tomorrow." He smiles and replies: "Then let's make it count." Medium shot, then a close-up on their handshake. Sound: office ambience, soft music.
Deepnia’s other video models, to pick the right one for your project.
| Model | Length | Resolution | Aspect ratios | Best for |
|---|---|---|---|---|
| Kling 3.0This model | 3–15 s | Set by the mode | 9:16 · 1:1 · 16:9 | Multi-shot AI video with your characters and sound. |
| Kling 2.6 | 5 or 10 s | Automatic | 9:16 · 1:1 · 16:9 | A simple 5 or 10 s clip from text or one image, with voices, singing and sound effects. |
| Seedance 2.0 | 4–15 s | 480p–1080p | 16:9 · 9:16 · 1:1 · 3:4 · 4:3 · 21:9 | Videos with sound up to 1080p, guided by image and video references. |
| Seedance 2.5 | 4–30 s | 480p–720p | 16:9 · 9:16 · 1:1 · 3:4 · 4:3 · 21:9 | Videos up to 30 s with up to 16 references. |
| Gemini Omni | 4, 6, 8 or 10 s | 720p–4K | 9:16 · 16:9 | 4–10 s clips up to 4K with 7 reference images. |
| Veo | 8 s | Set by the mode | 9:16 · 16:9 | 8 s clips in three quality levels. |
Up to 5 shots on Deepnia, each 1 to 12 seconds long, 15 seconds in total. Each shot has its own prompt of up to 500 characters.
A character or an object you define with 2–4 images or a video, then call in the prompt with @element_1, @element_2, @element_3 to keep it identical from shot to shot. You can use up to 3 per video.
Yes: turn on “Generated sound” and Kling 3.0 creates voices, dialogue, sound effects and ambience together with the picture. Describe the sound you want at the end of the prompt.
Turn on “Generated sound”, put each line in quotation marks and say who speaks, in what tone and in which language. Kling documents dialogue in Chinese, English, Japanese, Korean and Spanish, with Chinese dialects and English accents.
From 3 to 15 seconds, 5 seconds by default. In multi-shot mode, the total length is the sum of the shots.
They are Kling 3.0’s two generation modes on Deepnia. According to Kling’s documentation, Standard produces 720p video and Pro 1080p: test your shots in Standard, then switch to Pro for the final cut.
Yes, in single-shot mode: add a start frame, then an end frame if you like; Kling 3.0 animates the move from one to the other. In multi-shot mode, the start frame opens the video.
Kling 3.0 to tell a story over several shots, keep characters identical or pick any length from 3 to 15 seconds. Kling 2.6 for a simple 5 or 10 second clip from text or one image.
The video tool opens with Kling 3.0 already selected. The exact cost shows before every creation; you sign up when you generate.
