Skip to content
Get started
GENERATION API CALLS
Image Model Generation

BFL

This page is auto-generated from model configurations. Last updated: 2026-07-01.

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


Flux Kontext for advanced image editing and generation

Model ID: model_flux-kontext-editing

Capabilities: img2img

LLM Markdown: https://app.scenario.com/api/models/model_flux-kontext-editing/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
lorasmodel_arrayNo---List of one or more LoRA model IDs or URLs.
lorasScalenumber_arrayNo-02-Scales for the LoRA weights
referenceImagesfile_arrayYes----Reference images for style or content guidance
promptstringYes----Text prompt for image editing or generation
numOutputsnumberNo114-Number of images to generate
aspectRatiostringNoauto--21:9, 16:9, 3:2, 4:3, 1:1, 2:3, 3:4, 9:16, 9:21, autoAspect ratio for the generated image
qualitystringNohigh--high, mediumGeneration quality
guidanceScalenumberNo3.5110-How closely to follow the prompt (1.0-10.0)
seednumberNo-02147483647-Use a seed for reproducible results (single-image inference). Leave blank to use a random seed.

FLUX.1 Dev - High-quality text-to-image generation model

Model ID: model_bfl-flux-1-dev

Capabilities: txt2img, img2img, controlnet

LLM Markdown: https://app.scenario.com/api/models/model_bfl-flux-1-dev/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
modelIdmodelNo----A single LoRA or composition model ID. If a composition is provided, the loras and lorasScale arrays will be automatically populated from the composition’s concepts. If a LoRA is provided, it will be added to the loras array.
lorasmodel_arrayNo---List of one or more LoRA model IDs. If provided, this will override the LoRAs extracted from the composition (when modelId is a composition). If modelId is a LoRA and this is not provided, it will be automatically populated with the modelId.
lorasScalenumber_arrayNo-02-Scales for the LoRA weights
promptstringYes----Text prompt for image generation
imagefileNo----Input image for img2img
strengthnumberNo0.80.011-Lower values adhere more closely to the image, while higher values allow for more creative freedom. (Higher values are recommended for this model)
controlImagefileNo----Control image used to condition the generation (edge / depth / pose / etc., per the selected modality).
controlModalitystringNocanny--canny, tile, depth, blur, pose, gray, low-qualityHow the control image should be interpreted.
controlStrengthnumberNo0.701-How strongly the control image conditions the generation. Higher = output adheres more closely to the control image’s structure; lower = the prompt dominates. Recommended range 0.3–0.8. Starting points: canny / depth / tile ~0.7, pose / gray / blur / low-quality ~0.8–0.9. Above 0.9 can become rigid or artifact-prone.
controlStartnumberNo001-Fraction of diffusion steps (0–1) at which the control image begins influencing the generation. 0 = active from the first step. Strength is how much; Start / End is when.
controlEndnumberNo101-Fraction of diffusion steps (0–1) at which the control image stops influencing the generation. 1 = active until the last step. Lowering it (e.g. 0.65) locks composition early then releases so the prompt can refine details — useful for pose at high strength.
numOutputsnumberNo114-Number of images to generate.
numInferenceStepsnumberNo28150-The number of denoising steps
widthnumberNo11041282048-The width of the generated images, must be a multiple of 16
heightnumberNo8321282048-The height of the generated images, must be a multiple of 16
guidancenumberNo3.5010-Controls how closely the generated image follows the prompt (ignored for schnell)
seednumberNo-02147483647-Used to reproduce previous results

An opinionated text-to-image model from Black Forest Labs in collaboration with Krea that excels in photorealism. Creates images that avoid the oversaturated ‘AI look’.

Model ID: model_flux-krea-dev

Capabilities: txt2img, img2img

LLM Markdown: https://app.scenario.com/api/models/model_flux-krea-dev/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
promptstringNo----Describe your image
aspectRatiostringNo4:3--9:21, 9:16, 2:3, 3:4, 4:5, 1:1, 3:2, 4:3, 5:4, 16:9, 21:9Image aspect ratio. Ignored if you have a reference image.
imagefileNo----Use a reference image to guide your generation. The aspect ratio of your output will match this image
promptStrengthnumberNo0.801-Lower values adhere more closely to the image, while higher values allow for more creative freedom. (Higher values are recommended for this model)
numOutputsnumberNo114-Number of images to generate.
stepsnumberNo28150-The number of steps for the generation
guidancenumberNo3010-Higher values adhere more closely to the prompt, while lower values allow for more creative freedom.
megapixelsstringNo1--1, 0.25Approximate number of megapixels for generated image
seednumberNo-02147483647-Use a seed for reproducible results. Leave blank to use a random seed.

FLUX.1 Schnell - Fast text-to-image generation model

Model ID: model_bfl-flux-1-schnell

Capabilities: txt2img, img2img

LLM Markdown: https://app.scenario.com/api/models/model_bfl-flux-1-schnell/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
modelIdmodelNo----A single LoRA or composition model ID. If a composition is provided, the loras and lorasScale arrays will be automatically populated from the composition’s concepts. If a LoRA is provided, it will be added to the loras array.
lorasmodel_arrayNo---List of one or more LoRA model IDs. If provided, this will override the LoRAs extracted from the composition (when modelId is a composition). If modelId is a LoRA and this is not provided, it will be automatically populated with the modelId.
lorasScalenumber_arrayNo-02-Scales for the LoRA weights, overriding the composition scales.
promptstringYes----Text prompt for image generation
imagefileNo----Reference image
strengthnumberNo0.80.011-Lower values adhere more closely to the image, while higher values allow for more creative freedom. (Higher values are recommended for this model)
numOutputsnumberNo114-Number of images to generate.
numInferenceStepsnumberNo4110-The number of denoising steps
widthnumberNo11041282048-The width of the generated images, must be a multiple of 16
heightnumberNo8321282048-The height of the generated images, must be a multiple of 16
guidancenumberNo3.5010-Controls how closely the generated image follows the prompt (ignored for schnell)
seednumberNo-02147483647-Used to reproduce previous results

Faster, better FLUX Pro. Text-to-image model with excellent image quality, prompt adherence, and output diversity.

Model ID: model_bfl-flux-1-1-pro

Capabilities: txt2img

LLM Markdown: https://app.scenario.com/api/models/model_bfl-flux-1-1-pro/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
promptstringYes----Text prompt for image generation
imagePromptfileNo----Image to use with Flux Redux. This is used together with the text prompt to guide the generation towards the composition of the Image prompt.
aspectRatiostringNo1:1--16:9, 3:2, 4:3, 5:4, 1:1, 4:5, 3:4, 2:3, 9:16Aspect ratio for the generated image
promptUpsamplingbooleanNofalse---Automatically modify the prompt for more creative generation
seednumberNo----Random seed. Set for reproducible generation

FLUX1.1 [pro] in ultra and raw modes. Images are up to 4 megapixels. Use raw mode for realism.

Model ID: model_bfl-flux-1-1-pro-ultra

Capabilities: txt2img

LLM Markdown: https://app.scenario.com/api/models/model_bfl-flux-1-1-pro-ultra/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
promptstringYes----Text prompt for image generation
imagePromptfileNo----Image to use with Flux Redux. This is used together with the text prompt to guide the generation towards the composition of the image_prompt. Must be jpeg, png, gif, or webp.
aspectRatiostringNo1:1--21:9, 16:9, 3:2, 4:3, 5:4, 1:1, 4:5, 3:4, 2:3, 9:16, 9:21Aspect ratio for the generated image
rawbooleanNofalse---Generate less processed, more natural-looking images
seednumberNo----Random seed. Set for reproducible generation

FLUX.2 Dev - High-quality text-to-image and image-to-image model

Model ID: model_bfl-flux-2-dev

Capabilities: txt2img, img2img

LLM Markdown: https://app.scenario.com/api/models/model_bfl-flux-2-dev/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
modelIdmodelNo----A single LoRA model ID. If provided, it will be added to the loras array.
lorasmodel_arrayNo---List of one or more LoRA model IDs. If modelId is a LoRA and this is not provided, it will be automatically populated with the modelId.
lorasScalenumber_arrayNo-02-Scales for the LoRA weights
promptstringYes----Text prompt for image generation
referenceImagesfile_arrayNo----Reference images
numOutputsnumberNo114-Number of images to generate
numInferenceStepsnumberNo281100-The number of denoising steps
widthnumberNo-1282048-The width of the generated images, must be a multiple of 16
heightnumberNo-1282048-The height of the generated images, must be a multiple of 16
guidancenumberNo4110-Controls how closely the generated image follows the prompt
seednumberNo-02147483647-Used to reproduce previous results

FLUX.2 Flex is designed for maximum versatility and ease of use. It prioritizes adaptability, allowing fast generation, broad compatibility, and reliable results

Model ID: model_bfl-flux-2-flex-editing

Capabilities: img2img

LLM Markdown: https://app.scenario.com/api/models/model_bfl-flux-2-flex-editing/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
referenceImagesfile_arrayYes----Reference images for style or content guidance
promptstringYes----Text prompt for image generation
aspectRatiostringNo1:1--match_input_image, 16:9, 3:2, 4:3, 5:4, 1:1, 4:5, 3:4, 2:3, 9:16Aspect ratio for the generated image. When aspect ratio is ‘Match Input Image’, use ‘Match Input Image’ to match the input image’s aspect ratio.
resolutionstringNo1 MP--match_input_image, 0.5 MP, 1 MP, 2 MP, 4 MPResolution in megapixels. When aspect ratio is ‘Auto (Match Input)’, use ‘Auto (Match Input)’ to match the input image’s resolution (clamped to 0.5-4 MP).
promptUpsamplingbooleanNofalse---Automatically modify the prompt for more creative generation
stepsnumberNo30150-Number of steps to run the model for. Fewer steps means faster generation, at the expensive of output quality. 30 steps is sufficient for most prompts.
guidancenumberNo4.51.510-Guidance scale. Higher values mean more guidance, lower values mean more creative generation.
seednumberNo----Random seed. Set for reproducible generation

FLUX.2 Klein 4b - Efficient text-to-image and image-to-image model

Model ID: model_bfl-flux-2-klein-4b

Capabilities: txt2img, img2img

LLM Markdown: https://app.scenario.com/api/models/model_bfl-flux-2-klein-4b/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
modelIdmodelNo----A single LoRA model ID. If provided, it will be added to the loras array.
lorasmodel_arrayNo---List of one or more LoRA model IDs. If modelId is a LoRA and this is not provided, it will be automatically populated with the modelId.
lorasScalenumber_arrayNo-02-Scales for the LoRA weights
promptstringYes----Text prompt for image generation
referenceImagesfile_arrayNo----Reference images
numOutputsnumberNo114-Number of images to generate
numInferenceStepsnumberNo41100-The number of denoising steps
widthnumberNo-1282048-The width of the generated images, must be a multiple of 16
heightnumberNo-1282048-The height of the generated images, must be a multiple of 16
guidancenumberNo1110-Controls how closely the generated image follows the prompt
seednumberNo-02147483647-Used to reproduce previous results

FLUX.2 Klein 9b - Efficient text-to-image and image-to-image model

Model ID: model_bfl-flux-2-klein-9b

Capabilities: txt2img, img2img

LLM Markdown: https://app.scenario.com/api/models/model_bfl-flux-2-klein-9b/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
modelIdmodelNo----A single LoRA model ID. If provided, it will be added to the loras array.
lorasmodel_arrayNo---List of one or more LoRA model IDs. If modelId is a LoRA and this is not provided, it will be automatically populated with the modelId.
lorasScalenumber_arrayNo-02-Scales for the LoRA weights
promptstringYes----Text prompt for image generation
referenceImagesfile_arrayNo----Reference images for img2img
numOutputsnumberNo114-Number of images to generate
numInferenceStepsnumberNo41100-The number of denoising steps
widthnumberNo-1282048-The width of the generated images, must be a multiple of 16
heightnumberNo-1282048-The height of the generated images, must be a multiple of 16
guidancenumberNo1110-Controls how closely the generated image follows the prompt
seednumberNo-02147483647-Used to reproduce previous results

FLUX.2 Klein Base 4b - Efficient text-to-image and image-to-image model

Model ID: model_bfl-flux-2-klein-4b-base

Capabilities: txt2img, img2img

LLM Markdown: https://app.scenario.com/api/models/model_bfl-flux-2-klein-4b-base/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
modelIdmodelNo----A single LoRA model ID. If provided, it will be added to the loras array.
lorasmodel_arrayNo---List of one or more LoRA model IDs. If modelId is a LoRA and this is not provided, it will be automatically populated with the modelId.
lorasScalenumber_arrayNo-02-Scales for the LoRA weights
promptstringYes----Text prompt for image generation
referenceImagesfile_arrayNo----Reference images
numOutputsnumberNo114-Number of images to generate
numInferenceStepsnumberNo281100-The number of denoising steps
widthnumberNo-1282048-The width of the generated images, must be a multiple of 16
heightnumberNo-1282048-The height of the generated images, must be a multiple of 16
guidancenumberNo4110-Controls how closely the generated image follows the prompt
seednumberNo-02147483647-Used to reproduce previous results

FLUX.2 Klein Base 9b - Efficient text-to-image and image-to-image model

Model ID: model_bfl-flux-2-klein-9b-base

Capabilities: txt2img, img2img

LLM Markdown: https://app.scenario.com/api/models/model_bfl-flux-2-klein-9b-base/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
modelIdmodelNo----A single LoRA model ID. If provided, it will be added to the loras array.
lorasmodel_arrayNo---List of one or more LoRA model IDs. If modelId is a LoRA and this is not provided, it will be automatically populated with the modelId.
lorasScalenumber_arrayNo-02-Scales for the LoRA weights
promptstringYes----Text prompt for image generation
referenceImagesfile_arrayNo----Reference images for img2img
numOutputsnumberNo114-Number of images to generate
numInferenceStepsnumberNo281100-The number of denoising steps
widthnumberNo-1282048-The width of the generated images, must be a multiple of 16
heightnumberNo-1282048-The height of the generated images, must be a multiple of 16
guidancenumberNo4110-Controls how closely the generated image follows the prompt
seednumberNo-02147483647-Used to reproduce previous results

Model ID: model_bfl-flux-2-max-editing

Capabilities: txt2img, img2img

LLM Markdown: https://app.scenario.com/api/models/model_bfl-flux-2-max-editing/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
referenceImagesfile_arrayNo----Reference images for style or content guidance
promptstringYes----Text prompt for image generation
aspectRatiostringNo4:3--match_input_image, 16:9, 3:2, 4:3, 5:4, 1:1, 4:5, 3:4, 2:3, 9:16Aspect ratio for the generated image. Use ‘Auto (Match Input)’ to match the first input image’s aspect ratio.
resolutionstringNo1 MP--match_input_image, 0.5 MP, 1 MP, 2 MP, 4 MPResolution in megapixels. Up to 4 MP is possible, but 2 MP or below is recommended. The maximum image size is 2048x2048, which means that high-resolution images may not respect the resolution if aspect ratio is not 1:1. When Aspect ratio is ‘Auto (Match Input)’, use Auto to match the input image’s resolution (clamped to 0.5-4 MP).
seednumberNo----Random seed. Set for reproducible generation

FLUX.2 Pro is the highest-fidelity, flagship version of the FLUX.2 lineup

Model ID: model_bfl-flux-2-pro-editing

Capabilities: txt2img, img2img

LLM Markdown: https://app.scenario.com/api/models/model_bfl-flux-2-pro-editing/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
referenceImagesfile_arrayNo----Reference images for style or content guidance
promptstringYes----Text prompt for image generation
aspectRatiostringNo1:1--match_input_image, 16:9, 3:2, 4:3, 5:4, 1:1, 4:5, 3:4, 2:3, 9:16Aspect ratio for the generated image. When aspect ratio is ‘Match Input Image’, use ‘Match Input Image’ to match the input image’s aspect ratio.
resolutionstringNo1 MP--match_input_image, 0.5 MP, 1 MP, 2 MP, 4 MPResolution in megapixels. When aspect ratio is ‘Auto (Match Input)’, use ‘Auto (Match Input)’ to match the input image’s resolution (clamped to 0.5-4 MP).
promptUpsamplingbooleanNofalse---Automatically modify the prompt for more creative generation
seednumberNo----Random seed. Set for reproducible generation

Image-to-image editing with FLUX.2 [turbo] from Black Forest Labs. Precise modifications using natural language descriptions and hex color control—all at turbo speed.

Model ID: model_bfl-flux-2-turbo-editing

Capabilities: img2img

LLM Markdown: https://app.scenario.com/api/models/model_bfl-flux-2-turbo-editing/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
referenceImagesfile_arrayYes----Reference images for style or content guidance
promptstringYes----Text prompt for image generation
imageSizestringNolandscape_4_3--portrait_16_9, portrait_4_3, square_hd, landscape_4_3, landscape_16_9Image size for the generated image.
guidancenumberNo2.51.510-Guidance scale. Higher values mean more guidance, lower values mean more creative generation.
numOutputsnumberNo114-Number of images to generate.
promptUpsamplingbooleanNofalse---Automatically modify the prompt for more creative generation
seednumberNo----Random seed. Set for reproducible generation

Generates a seam-continuous equirectangular 360° panorama from a text prompt, powered by FLUX.1 Dev with the equirectangular geometry LoRA and black-band removal applied automatically.

Model ID: model_scenario-skybox-flux

Capabilities: txt2img, img2img

LLM Markdown: https://app.scenario.com/api/models/model_scenario-skybox-flux/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
promptstringYes----Describe the scene you want to see all around you — a sunset beach, a futuristic city, a forest at dawn. The model automatically formats your prompt to produce a seamless 360° panorama.
stylestringNocinematic--standard, cinematic, claymation, cloud-skydome, cyberpunk, fantasy, neon-tron, photo, psychedelic, retro-fantasy, scifi-concept-art, space, whimsical, 3d-cartoon, cartoon, comic, manga, ink, pastel, oil-painting, enchanted, manga-colorThe artistic look of your panorama. Choose from cinematic, fantasy, cyberpunk, oil painting, and more. Each style shapes the color palette, lighting, and overall mood. Use Standard for a neutral, unstyled result.
negativePromptstringNo----Describe what you don’t want in the image — e.g. ‘blurry, people, text’. Has no effect unless Negative Prompt Strength is above 0. Try generating without one first to see if it’s needed.
negativePromptStrengthnumberNo0010-How strongly the model avoids what you described in the Negative Prompt. Leave at 0 to disable it entirely. Increase gradually — values between 1–4 are usually enough.
imagefileNo----Upload an existing image to use as a starting point. The model will generate a new panorama inspired by it, with the amount of change controlled by the Strength slider below.
strengthnumberNo0.80.011-How much the model transforms your reference image. Low values stay close to the original; high values produce a more creative, reimagined result.
numOutputsnumberNo214-How many panoramas to generate at once. Useful for exploring variations before committing to one.
numInferenceStepsnumberNo302050-Number of refinement passes the model takes. More steps generally produce sharper, more detailed results but take longer. 30 is a good default.
geometryEnforcementnumberNo00100-Fine-tunes how strongly the model enforces correct 360° geometry. Leave at 0 to use the recommended level for your chosen style. Increase only if the panorama seam or poles look distorted.
seednumberNo-02147483647-A number that locks in the randomness of the generation. Copy the seed from a result you liked to reproduce it exactly, or leave blank for a new result each time.

Model ID: model_sc-upscale-flux-skybox

Capabilities: img2img

LLM Markdown: https://app.scenario.com/api/models/model_sc-upscale-flux-skybox/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
imagefileYes----Image to upscale
upscaleFactornumberNo228-Upscale factor
promptstringNo----Prompt
imagePromptfile_arrayNo----Upload up to 4 images to guide the style of the output.
imagePromptStrengthnumberNo0.601-Higher fidelity keeps the enhanced image’s style closely aligned with the style images
controlnetConditioningScalenumberNo0.501-Higher values better preserve the structure of the input image.
strengthnumberNo0.601-Higher values introduce more imaginative elements to the output image.
numInferenceStepsnumberNo28150-The number of steps for the generation
detailsEnablebooleanNotrue---Enhances fine details in the output image.
baseModelstringNoFLUX.1-dev--FLUX.1-dev, FLUX.1-Krea-devUse FLUX.1-dev for stylized images, and FLUX.1-Krea-dev for realistic images.
seednumberNo-02147483647-Seed used for reproduction

Model ID: model_sc-upscale-flux-texture

Capabilities: img2img

LLM Markdown: https://app.scenario.com/api/models/model_sc-upscale-flux-texture/markdown

ParameterTypeRequiredDefaultMinMaxAllowed ValuesDescription
imagefileYes----Image to upscale
upscaleFactornumberNo228-Upscale factor
presetstringNobalanced--precise, balanced, creativeSelect a preset to apply a set of parameters
lorasmodel_arrayNo---List of one or more LoRA model IDs or URLs.
lorasScalenumber_arrayNo---Scales for the LoRA weights
promptstringNo----Prompt
imagePromptfile_arrayNo----Upload up to 4 images to guide the style of the output.
imagePromptStrengthnumberNo0.601-Higher fidelity keeps the enhanced image’s style closely aligned with the style images
controlnetConditioningScalenumberNo0.401-Higher values better preserve the structure of the input image.
strengthnumberNo0.401-Higher values introduce more imaginative elements to the output image.
numInferenceStepsnumberNo28150-The number of steps for the generation
detailsEnablebooleanNotrue---Enhances fine details in the output image.
baseModelstringNoFLUX.1-dev--FLUX.1-dev, FLUX.1-Krea-devUse FLUX.1-dev for stylized images, and FLUX.1-Krea-dev for realistic images.
seednumberNo-02147483647-Seed used for reproduction