23 models live from Kling, Sonilo and BytePlus, one API key

23 models live · 3 providers · one API

One API key.
Every video, music and audio model.

Call 23 generative models from Kling, Sonilo and BytePlus through one HTTP/JSON API. Every model has a published price, usage comes out of one prepaid wallet, and there is no subscription.

Get API key Browse models

Sign up with Google or email · top up from $10 · pay per request

Dreamina Seedance 2.0BytePlus · Provider costDreamina Seedance 2.0 FastBytePlus · Provider costDreamina Seedance 2.0 MiniBytePlus · Provider costDreamina Seedance 2.5BytePlus · Provider costKling 3.0 Omni Pro Video EditKling · ≈ $0.84Kling 3.0 Omni Pro Video ReferenceKling · ≈ $0.84Kling 3.0 Turbo Pro Image to VideoKling · ≈ $0.70Kling 3.0 Turbo Pro Text to VideoKling · ≈ $0.70Kling 3.0 Turbo Standard Image to VideoKling · ≈ $0.56Kling 3.0 Turbo Standard Text to VideoKling · ≈ $0.56Kling Video O3 4k Image to VideoKling · ≈ $2.10Kling Video O3 4k Reference to VideoKling · ≈ $2.10Kling Video O3 4k Text to VideoKling · ≈ $2.10Kling Video V3 4k Image to VideoKling · ≈ $2.10Kling Video V3 4k Text to VideoKling · ≈ $2.10Kling Video V3 Pro Image to VideoKling · ≈ $0.84Kling Video V3 Pro Motion ControlKling · ≈ $0.84Kling Video V3 StandardKling · ≈ $0.63SOSonilo Video DubbingSonilo · $0.0985/sSOSonilo Text to MusicSonilo · $0.00225/sSOSonilo Text to Sound EffectsSonilo · $0.0018/sSOSonilo Video to MusicSonilo · $0.009/sSOSonilo Video to Sound EffectsSonilo · $0.009/s

One request shape

Switch models by changing one line

Every media model takes the same request: a model ID and its params. Submit, poll, download. Pick a type to see a real request and what it costs.

RequestPOST /api/media/generate
{
  "model": "kwaivgi/kling-video/3.0/turbo/standard/image-to-video",
  "params": {
    "image": "https://example.com/your-image.png",
    "prompt": "He sways gently into the note, cheeks pressing, sweat glistening, smoke drifting through the warm light. The camera eases in slowly. Photorealistic, soulful, moody.",
    "duration": 5
  }
}
Job lifecycle
  1. Submitted with your apiKeycode 0
  2. Returns a jobId straight awaypending
  3. Poll /api/media/resultpending
  4. Output URL in resultDatacompleted
Cost
Billed at the provider's actual cost; a typical clip is listed at $0.56. Your usage log shows the exact charge.
Same request shape for every model

One API key

Create keys in the console and use the same key for every model. No provider accounts to open.

One request format

Consistent apiKey header, code/data/msg envelope and job API across providers.

One prepaid wallet

Top up from $10. Every charge is in your transaction history and usage log.

Provider fallback

Where a model has several providers configured, retryable failures move to the next one.

Browse by task

Start from what you want to make

Each task page compares the models that can do it, with prices and inputs side by side.

Catalog · 23 models · one key

Every model.
One key.

Models from Kling, Sonilo and BytePlus, each with its own page: parameters, request examples, price and billing unit. New models appear in the catalog as soon as they are enabled.

Browse all models See pricing

Why ScalingTensor?

Built for developers who want to ship media features without juggling vendors.

Agent-ready, via SKILL.md

Point Claude Code, Codex or any coding agent at our SKILL.md and it learns how to pick a model, submit a job and poll the result.

$ claude "Set up ScalingTensor from
  https://scalingtensor.com/SKILL.md"

✓ reads the model catalog
✓ asks you for SCALINGTENSOR_API_KEY
✓ submits a job and polls the result

Async jobs, done properly

Video and audio take seconds to minutes. Submit returns a jobId at once; poll it from your backend, and results come back with the output URL.

Read the polling guide

Every charge on record

One wallet for every model. The console shows your balance, each top-up and charge, and a usage log per request.

Prepaid walletUsage logTop-ups from $10No subscription
API FAQ

Frequently asked questions

What is ScalingTensor?

ScalingTensor is an API marketplace for AI models. You call every model in the catalog through one HTTP/JSON API with a single API key, and pay per request from a prepaid wallet.

Where can I find available models and API pricing?

The model catalog lists every enabled model with its configured price per job, per second, per token or at the provider’s actual cost. A catalog listing does not by itself guarantee that a provider adapter is configured, so test a model before depending on it.

Can I use the OpenAI SDK?

Not directly. Media models use POST /api/media/generate and GET /api/media/result, and the chat endpoint is POST /api/chat/completion. All of them take an apiKey header and return code and data rather than the OpenAI SDK format, so an adapter is required.

How does pricing work?

Each model page shows how it is billed: per job, per second of audio or video, per output quantity, per token, or at the upstream provider’s actual cost. Charges are deducted from your prepaid wallet.

How do I create an account?

Sign up with Google or your email address on the sign-up page; email sign-ups confirm the address with a one-time code. You can then create API keys and top up your wallet.

Is my data secure?

Use HTTPS and keep API keys on your server. Model providers may have different data handling policies; review those policies before submitting sensitive content.

Support

Questions? Talk to us.

Choosing a model, estimating cost at volume, or getting a first request working: email us, or start from the docs.

Build with 23 models today

Sign up with Google or email, top up from $10 and send your first request in minutes.

Get API key Contact sales