Krea 2 Medium
Fast, expressive text-to-image generation from Krea's foundation model, tuned for illustration, anime, and painterly styles. Krea 2 Medium supports up to 10 style-reference images that transfer a look — not content — onto your output, plus a four-level creativity control from literal 'raw' to free-wheeling 'high'.
Optional reference images (up to 10) whose style is transferred to the output. They guide the look, not the content. Adding any costs slightly more.
Aspect Ratio
Aspect ratio of the generated image.
Creativity
How loosely the model interprets the prompt. 'raw' renders only what you describe, 'high' takes real creative liberty.
Your result renders here.
Use Krea 2 Medium via the API
Generate with Krea 2 Medium from your own code. Async by design: submit a generation, get a run id back instantly, poll until it completes. Billed in credits at the best available provider price — from $0.03 per image.
1 — Start a generation
curl -X POST https://app.scenetra.com/api/v1/generate/krea-2-medium \
-H "Authorization: Bearer YOUR_API_KEY" \
-H "Content-Type: application/json" \
-d '{"prompt":"a watercolor fox in falling snow, soft light","aspect_ratio":"1:1","creativity":"medium"}'Response (202)
{
"runId": "run_abc123",
"status": "PENDING",
"model": "krea-2-medium",
"estimatedCredits": 30,
"statusUrl": "/api/v1/runs/run_abc123"
}2 — Poll until COMPLETED (every 2-5 seconds)
curl https://app.scenetra.com/api/v1/runs/run_abc123 \
-H "Authorization: Bearer YOUR_API_KEY"Response
{
"runId": "run_abc123",
"status": "COMPLETED",
"progress": { "completed": 1, "total": 1 },
"outputs": [
{ "type": "image", "url": "https://images.scenetra.com/..." }
],
"error": null,
"estimatedCredits": 30
}Request fields
| Field | Type | Description | Default |
|---|---|---|---|
| prompt* | string | Describe the image to generate. | — |
| style_reference_images | string[] (https URLs) | Optional reference images (up to 10) whose style is transferred to the output. They guide the look, not the content. Adding any costs slightly more. | — |
| aspect_ratio | string | Aspect ratio of the generated image.1:1 · 4:3 · 3:2 · 16:9 · 2.35:1 · 4:5 · 2:3 · 9:16 | 1:1 |
| creativity | string | How loosely the model interprets the prompt. 'raw' renders only what you describe, 'high' takes real creative liberty.raw · low · medium · high | medium |
| callbackUrl | string (https URL) | Optional. When the run finishes (completed or failed), we POST the final run status to this URL — same shape as the poll response, plus model, params, and chargedCredits. Best-effort; polling remains the source of truth. | |
Unknown fields are rejected with a helpful 400. Media fields take https URLs. There is no provider parameter — every run is routed to the cheapest capable provider automatically.
Good to know
Billing
Runs bill your account credits at provider cost with 0% markup. The submit response includes the credit estimate; failed generations are refunded automatically. Add ?dryRun=1 to get the estimate without running anything.
Errors
Errors are always { "error": { "code", "message" } } with stable codes: unauthorized, model_not_available, invalid_parameter, insufficient_credits, rate_limited. A failed run is not an HTTP error — poll responses report status: "FAILED" with the reason.
What makes Krea 2 Medium different
Tuned for illustration, anime, and paint
Where most foundation models chase photorealism first, Krea 2 Medium is tuned for expressive styles — illustration, anime, and painterly rendering with real character. If your output is meant to look drawn or painted rather than photographed, Medium is the tier built for it; when you need photorealism, Krea 2 Large sits one node away.
Style references that transfer look, not content
Attach up to 10 style-reference images and Krea 2 Medium absorbs their aesthetic — palette, linework, rendering technique — while the prompt still controls what's in the frame. It's the clean way to keep a series visually consistent: same references, different prompts, one coherent body of work. Adding references costs $0.005 extra per image.
A creativity dial from literal to liberated
Four interpretation levels change how much liberty the model takes: 'raw' renders only what you describe, 'low' and 'medium' balance fidelity and flair, and 'high' treats your prompt as a starting point for its own ideas. Exploration runs want 'high'; production runs against a tight brief want 'raw'.
Cheap enough to iterate freely
At $0.03 per image ($0.035 with style references), a dollar buys about 33 generations — and it's billed at provider cost with 0% markup. On a Scenetra board you can fan one prompt across several creativity levels or reference sets and compare the results side by side.
Parameters
| Parameter | Type | Description | Default |
|---|---|---|---|
| Prompt* | text | Describe the image to generate. | — |
| Style References | image upload | Optional reference images (up to 10) whose style is transferred to the output — they guide the look, not the content. Adding any costs slightly more. | — |
| Aspect Ratio | select | Aspect ratio of the generated image.1:1 · 4:3 · 3:2 · 16:9 · 2.35:1 · 4:5 · 2:3 · 9:16 | 1:1 |
| Creativity | select | How loosely the model interprets the prompt — 'raw' renders only what you describe, 'high' takes real creative liberty.raw · low · medium · high | medium |
Pricing
From $0.03 per image
Pay per generation — only pay for what you use.
Text-to-image
$0.03
per image
With style references
$0.035
per image
Use cases
Illustration & comics
- Character art in a consistent style
- Anime frames and covers
- Panel exploration from one style set
- Painterly key art
Branding & editorial
- Illustrated brand imagery from style refs
- Editorial spot illustrations
- Consistent blog art series
- Stylized icons and headers
Social content
- Stylized posts in 4:5 and 9:16
- Series with one signature look
- Cinematic 2.35:1 visuals
- Fast trend art at low cost
Game & concept art
- Style exploration with the creativity dial
- Environment concepts
- Stylized asset ideation
- Moodboards from reference sets
Frequently asked questions
What is Krea 2 Medium?+
Krea 2 Medium is the fast, expressive tier of Krea's foundation image model, tuned for illustration, anime, and painterly styles. It generates images from text prompts, optionally guided by up to 10 style-reference images and a four-level creativity control. On Scenetra it runs as a node in the visual workflow editor, alongside Krea 2 Large.
How much does Krea 2 Medium cost?+
$0.03 per image on Scenetra, or $0.035 when you attach style-reference images. Billing is per generation at provider cost with 0% markup, from your credit balance — no subscription required.
What do style references do?+
Style references transfer the look of your images — palette, linework, rendering style — onto the generated output, while your prompt keeps control of the content. You can attach up to 10. They are not an editing input: Krea 2 Medium does not modify the reference images themselves.
Can Krea 2 Medium edit existing images?+
No — Krea 2 Medium is text-to-image only. Its reference images guide style, not content, and there is no edit mode. If you need image editing, models like Nano Banana 2 or Seedream 5 Pro on the same Scenetra board handle that step, and you can chain them with Krea nodes in one workflow.
What does the creativity setting change?+
It controls how loosely the model interprets your prompt. 'Raw' renders strictly what you describe, 'low' and 'medium' add measured flair, and 'high' takes real creative liberty. Use 'raw' for tight briefs and 'high' for exploration.
How is Krea 2 Medium different from Krea 2 Large?+
Medium is the fast, affordable tier tuned for illustration and expressive styles at $0.03 per image. Large is Krea's flagship, strongest at photorealism and rich detail, at $0.06 per image. Both take style references and the same creativity control, so switching tiers is just swapping the node.
Related models
Krea 2 Large
Krea's flagship foundation image model, strongest at photorealism and rich, expressive styles. Krea 2 Large generates from text prompts with up to 10 style-reference images that transfer a look onto the output, a four-level creativity control, and eight aspect ratios up to cinematic 2.35:1.
View model →Flux 2 Pro
Black Forest Labs' professional-grade image model delivering high-quality outputs with precise control. Flux 2 Pro supports text-to-image generation and image editing with flexible sizing, safety controls, and simple flat per-image pricing.
View model →Nano Banana 2
Generate high-quality images at up to 4K resolution using Google's Gemini 3.1 Flash model. Nano Banana 2 is optimized for speed, supports text-to-image and image editing with reference images, and offers flexible aspect ratios.
View model →Start creating with Krea 2 Medium
Use Krea 2 Medium alongside 50+ other AI models in Scenetra's visual workflow editor. No setup required.
Get Started