Skip to content
Get started
GENERATION API CALLS
Image Model Generation

Krea

This page is auto-generated from model configurations. Last updated: 2026-11-05.

This reference lists all available Krea image generation models and their parameters. Use these parameter names when calling the Generation API.


Generate high-fidelity images from text with Krea 2 Large, with aspect ratio, creativity, seed, and optional style references.

Model ID: model_krea-2-large

Capabilities: txt2img

LLM Markdown: https://app.scenario.com/api/models/model_krea-2-large/markdown

Parameter Type Required Default Min Max Allowed Values Description
prompt string Yes - - - - Describe the image you want to generate, including subject, style, mood, and composition.
styleReferenceImages inputs_array No - - - - Optional reference images that guide the visual style of the generated image. Using style references slightly increases cost.
aspectRatio string No 1:1 - - 1:1, 4:3, 3:2, 16:9, 2.35:1, 4:5, 2:3, 9:16 Shape of the image, from square 1:1 to cinematic 2.35:1 widescreen or tall 9:16.
creativity string No medium - - raw, low, medium, high How loosely the model interprets the prompt. Higher values produce more creative results that may drift from the prompt; lower values stay closer to it.
seed number No - 0 2147483647 - Random seed for reproducible generation. Leave blank to use a random seed.

Generate high-quality images from text with Krea 2 Medium, with aspect ratio, creativity, seed, and optional style references.

Model ID: model_krea-2-medium

Capabilities: txt2img

LLM Markdown: https://app.scenario.com/api/models/model_krea-2-medium/markdown

Parameter Type Required Default Min Max Allowed Values Description
prompt string Yes - - - - Describe the image you want to generate, including subject, style, mood, and composition.
styleReferenceImages inputs_array No - - - - Optional reference images that guide the visual style of the generated image. Using style references slightly increases cost.
aspectRatio string No 1:1 - - 1:1, 4:3, 3:2, 16:9, 2.35:1, 4:5, 2:3, 9:16 Shape of the image, from square 1:1 to cinematic 2.35:1 widescreen or tall 9:16.
creativity string No medium - - raw, low, medium, high How loosely the model interprets the prompt. Higher values produce more creative results that may drift from the prompt; lower values stay closer to it.
seed number No - 0 2147483647 - Random seed for reproducible generation. Leave blank to use a random seed.

Generate high-fidelity images extremely fast from text with Krea 2 Medium Turbo, with aspect ratio, creativity, seed, and optional style references.

Model ID: model_krea-2-medium-turbo

Capabilities: txt2img

LLM Markdown: https://app.scenario.com/api/models/model_krea-2-medium-turbo/markdown

Parameter Type Required Default Min Max Allowed Values Description
prompt string Yes - - - - Describe the image you want to generate, including subject, style, mood, and composition.
styleReferenceImages inputs_array No - - - - Optional reference images that guide the visual style of the generated image. Using style references slightly increases cost.
aspectRatio string No 1:1 - - 1:1, 4:3, 3:2, 16:9, 2.35:1, 4:5, 2:3, 9:16 Shape of the image, from square 1:1 to cinematic 2.35:1 widescreen or tall 9:16.
creativity string No medium - - raw, low, medium, high How loosely the model interprets the prompt. Higher values produce more creative results that may drift from the prompt; lower values stay closer to it.
seed number No - 0 2147483647 - Random seed for reproducible generation. Leave blank to use a random seed.

Generate high-fidelity images from text in seconds with Krea 2 Turbo, the speed-optimized open-source version of Krea 2.

Model ID: model_krea-2-turbo

Capabilities: txt2img

LLM Markdown: https://app.scenario.com/api/models/model_krea-2-turbo/markdown

Parameter Type Required Default Min Max Allowed Values Description
prompt string Yes - - - - Describe the image you want to generate, including subject, style, mood, and composition.
imageSize string No square_hd - - portrait_16_9, portrait_4_3, square, square_hd, landscape_4_3, landscape_16_9 The size and shape of the generated image.
numOutputs number No 1 1 4 - How many images to generate from the same prompt. Each additional image adds to the cost.
acceleration string No none - - none, regular The acceleration level to use for image generation.
enablePromptExpansion boolean No false - - - When enabled, the prompt is expanded with an LLM into a more detailed image-generation prompt. On failure, the original prompt is used unchanged.
seed number No - 0 2147483647 - Random seed for reproducible generation. Image i uses seed + i. Leave blank to use a random seed.

Generate high-fidelity images from text with Krea 2 Turbo, guided by one to three style reference images.

Model ID: model_krea-2-turbo-style

Capabilities: txt2img

LLM Markdown: https://app.scenario.com/api/models/model_krea-2-turbo-style/markdown

Parameter Type Required Default Min Max Allowed Values Description
prompt string Yes - - - - Describe the image you want to generate, including subject, style, mood, and composition.
styleReferenceImages file_array Yes - - - - One to three images used as style references for the generation.
styleScale number No 1 0 4 - How strongly the style reference images influence the generated image.
imageSize string No square_hd - - portrait_16_9, portrait_4_3, square, square_hd, landscape_4_3, landscape_16_9 The size and shape of the generated image.
numOutputs number No 1 1 4 - How many images to generate from the same prompt. Each additional image adds to the cost.
acceleration string No none - - none, regular The acceleration level to use for image generation.
enablePromptExpansion boolean No false - - - When enabled, the prompt is expanded with an LLM into a more detailed image-generation prompt. On failure, the original prompt is used unchanged.
seed number No - 0 2147483647 - Random seed for reproducible generation. Image i uses seed + i. Leave blank to use a random seed.

Generate high-quality images from text with Krea 3, choosing between the Berry (quality) and Jam (speed) models.

Model ID: model_krea-3

Capabilities: txt2img

LLM Markdown: https://app.scenario.com/api/models/model_krea-3/markdown

Parameter Type Required Default Min Max Allowed Values Description
prompt string Yes - - - - Describe the image you want to generate, including subject, style, mood, and composition.
model string No berry - - berry, jam Berry favors quality; Jam favors speed.
imageSize string No auto - - auto, square_hd, square, portrait_4_3, portrait_16_9, landscape_4_3, landscape_16_9 Output size and aspect ratio. Auto lets the model pick the aspect ratio at about 2048x2048. Larger images cost more.
thinkingLevel string No low - - low, medium How much effort the model spends planning the prompt before generating. Higher levels are slower.
negativePrompt string No - - - - What to keep out of the image. Only used by the Berry model.
seed number No - 0 2147483647 - Random seed for reproducible generation. Leave blank to use a random seed.

Edit or combine up to eight images from text instructions with Krea 3, with an optional high reasoning level that plans the edit and keeps unedited areas intact.

Model ID: model_krea-3-edit

Capabilities: img2img

LLM Markdown: https://app.scenario.com/api/models/model_krea-3-edit/markdown

Parameter Type Required Default Min Max Allowed Values Description
prompt string Yes - - - - Describe the edit to apply. Refer to each reference image by its position, e.g. <image-1>, <image-2>.
referenceImages file_array Yes - - - - One to eight images to edit or combine. Their order sets the <image-N> tags used in the instructions. Each image is billed by its size, up to the output size.
model string No berry - - berry, jam Berry favors quality; Jam favors speed.
imageSize string No auto - - auto, square_hd, square, portrait_4_3, portrait_16_9, landscape_4_3, landscape_16_9 Output size and aspect ratio. Auto picks the aspect ratio at about 2048x2048. Larger images cost more.
thinkingLevel string No low - - low, medium, high How much effort the model spends planning the edit. Higher levels are slower. High plans the edit with segmentation, web search and reference photos, keeps unedited areas of a single input image, fixes required text, and costs more.
negativePrompt string No - - - - What to keep out of the image. Only used by the Berry model.
seed number No - 0 2147483647 - Random seed for reproducible generation. Use a new seed when editing a previous output again.