Pay only for what you generate
No monthly minimum. No seat fees. Pay-as-you-go pricing per generation
API billing is separate from Krea consumer plans
Already a Krea Pro/Max subscriber? That covers the Krea app only. API access is billed independently below.
Krea 2
K2 Medium Turbo
Best for fast iteration
- ~3s per generation
- 1.5K native (4K via Enhancer)
- Lowest cost per K2-quality image
$0.0150 price per image, USD
Start with TurboK2 Medium
Best for expressive illustrations
- ~10s per generation
- 1.5K native (4K via Enhancer)
- Balanced cost and quality
$0.0300 price per image, USD
Start with Medium Recommended
K2 Large
Best for photorealism
- ~25s per generation
- 2K native (4K via Enhancer)
- Highest fidelity K2 output
$0.0600 price per image, USD
Start with Large73 models
Image generation
Video generation
Generating 10K+ images / month?
Talk to us about volume discounts, dedicated capacity, and SLAs.
Frequently Asked Questions
API usage is billed per generation based on the compute time each request consumes. You only pay for what you generate — there are no monthly minimums, seat fees, or commitments. Pricing varies by model and is listed in the table above for every supported endpoint.
Yes. API access is billed independently from Krea Pro, Max, or any team plan. Your consumer subscription covers the Krea app and its included compute units; the API is metered against your API balance and invoiced separately.
Click "Get API key" at the top of this page, or open the API tokens page in your account settings. From there you can generate a token, top up your balance, and start making requests within a few minutes — every endpoint listed above works the moment your token is active.
If you're generating 100K+ images per month or running sustained production workloads, talk to our sales team about volume pricing, reserved capacity, and custom SLAs.
All output generated through the Krea API is licensed for commercial use, including marketing, products, and client work. A handful of partner-hosted models carry their own terms, which we surface in the model documentation when relevant.
Default rate and concurrency limits are tuned to support typical production workloads. You can see your current limits in your dashboard, and Enterprise customers can request elevated limits with dedicated capacity.
Every model listed in the table above is callable via the API. This includes Krea's K2 family alongside leading partner models for image and video generation. New models are added to the API on the same day they ship to the consumer app whenever possible.
Standard API access is best-effort with public uptime status. Enterprise plans add uptime SLAs, dedicated support, prioritized request routing, and the option of dedicated capacity for guaranteed throughput.