Describe a scene or upload a frame. A clip starts at $0.06, with no subscription.
A video model turns a text description into a clip. You write who is in the frame, what happens and how the camera moves. A 5 or 10 second clip is ready in a few minutes. You can also start from a photo, and the model will bring the frame to life.
The models differ a lot. Kling 3.0 keeps a character's face and clothes stable. Veo 3.1 adds sound and speech in the same pass. Seedance handles long scenes, up to 30 seconds in version 2.5. Hailuo is good at running, dancing and sports. On NeuralBox they all share one balance, so you can test one idea in several models.
You pay per clip, with no monthly plan. Tokens never expire. If a generation fails, the tokens go back to your balance.
Describe the subject, the action, the place and the camera move. The more specific, the closer the result.
| Models | Kling, Veo, Seedance, Hailuo, Wan, PixVerse, Runway, Grok Imagine and more |
| Modes | text to video, image to video, first and last frame, video from a reference |
| Length | usually 5 or 10 seconds, up to 30 seconds in Seedance 2.5 and Wan 3.0 |
| Quality | 480p to 4K, depends on the model |
| Aspect ratios | 16:9, 9:16, 1:1 and more |
| Clip | Tokens | USD |
|---|---|---|
| Hailuo 3.0, 6 s, 768p | 12,120 tokens | ≈ $0.40 |
| Seedance 1.5, 4 s, 480p, no sound | 1,750 tokens | ≈ $0.058 |
| Seedance 2.0 Mini, 5 s, 480p | 4,750 tokens | ≈ $0.16 |
| PixVerse V6, 5 s, 540p | 7,000 tokens | ≈ $0.23 |
| Veo 3.1 Lite, 720p, with sound | 7,500 tokens | ≈ $0.25 |
| Kling 3.0 Standard, 5 s | 17,500 tokens | ≈ $0.58 |
| Kling 3.0 Standard with sound, 5 s | 25,000 tokens | ≈ $0.83 |
USD shown at the Basic rate: $5 = 150,000 tokens. Larger plans cut the token price by up to 40%. Tokens never expire, and there is no subscription. The exact price shows before you start.
Open Video, pick a model and describe the scene. A clip is usually ready in a few minutes, and you can download it as MP4.
Yes. Upload an image and write what should happen. Kling, Veo, Seedance, Hailuo, Wan and PixVerse all support it.
It depends on the job. Take Veo 3.1 for a clip with sound. Kling 3.0 keeps a character consistent. Seedance suits long scenes. Seedance 1.5 and Seedance 2.0 Mini are inexpensive for tests.
A 4-second Seedance 1.5 clip at 480p without sound costs 1,750 tokens, which is ≈ $0.058. A 5-second Kling 3.0 Standard clip costs 17,500 tokens.
No. You top up once and pay per clip. Tokens never expire.