openapi: 3.0.0 servers: - url: https://prod.api.market/api/v1/magicapi/ai-image-variation-generator info: title: High-Quality Image Variation API (SDXL-Finetunes) description: Transform existing images into new creative variations with powerful SDXL-Finetunesfast, photorealistic, and perfect for visual content creation. version: 08807c2a252379c60ca5bc07e25d8f26fecfa9ad1d899987be8865e6d6a47ef2 components: schemas: HTTPValidationError: properties: detail: items: properties: loc: items: anyOf: - type: string - type: integer title: Location type: array msg: title: Message type: string type: title: Error Type type: string required: - loc - msg - type title: ValidationError type: object title: Detail type: array title: HTTPValidationError type: object Input: properties: cfg: default: 7.5 description: Guidance scale maximum: 20 minimum: 1 title: Cfg type: number x-order: 5 example: 2.47 checkpoint_name: default: realismEngineSDXL_v30VAE.safetensors description: Checkpoint model to use enum: - cyberrealisticXL_v5.safetensors - albedobaseXL_v31Large.safetensors - realDream_sdxl5.safetensors - realismEngineSDXL_v30VAE.safetensors - realvisxlV50_v50Bakedvae.safetensors title: checkpoint_name type: string x-order: 3 example: realvisxlV50_v50Bakedvae.safetensors controlnet_end_percent: default: 0.5 description: ControlNet end percentage maximum: 1 minimum: 0 title: Controlnet End Percent type: number x-order: 20 example: 0.5 controlnet_start_percent: default: 0 description: ControlNet start percentage maximum: 1 minimum: 0 title: Controlnet Start Percent type: number x-order: 19 example: 0 controlnet_strength: default: 1 description: ControlNet strength maximum: 2 minimum: 0 title: Controlnet Strength type: number x-order: 18 example: 1 denoise: default: 0.8 description: Denoising strength maximum: 1 minimum: 0 title: Denoise type: number x-order: 6 example: 0.8 height: default: 1024 description: Height of the image maximum: 2048 minimum: 512 title: Height type: integer x-order: 23 example: 2048 input_image: description: Input image for processing. If not provided, the default image will be used. format: uri title: Input Image type: string x-order: 0 example: https://replicate.delivery/pbxt/MkFZUXh7QUtpvY81ENJOEH0tueIRKIyn0jETUY8Q15RO29xe/DALL%C2%B7E%202023-10-27%2017.55.26%20-%20Lifestyle%20product%20photo%20showcasing%20a%20glistening%20perfume%20bottle%20on%20a%20wooden%20table%2C%20catching%20the%20ambient%20light.%20In%20the%20soft-focus%20background%2C%20a%20Swedish%20.png ipadapter_end_at: default: 1 description: IP-Adapter end at step percentage maximum: 1 minimum: 0 title: Ipadapter End At type: number x-order: 17 example: 1 ipadapter_start_at: default: 0 description: IP-Adapter start at step percentage maximum: 1 minimum: 0 title: Ipadapter Start At type: number x-order: 16 example: 0 ipadapter_weight: default: 0.9 description: IP-Adapter weight maximum: 1 minimum: 0 title: Ipadapter Weight type: number x-order: 15 example: 0.9 lora1_strength_clip: default: 1 description: LoRA 1 (product marketing) clip strength maximum: 2 minimum: 0 title: Lora1 Strength Clip type: number x-order: 10 example: 1 lora1_strength_model: default: 1 description: LoRA 1 (product marketing) model strength maximum: 2 minimum: 0 title: Lora1 Strength Model type: number x-order: 9 example: 0.5 lora2_strength_clip: default: 1 description: LoRA 2 (add-detail-xl) clip strength maximum: 2 minimum: 0 title: Lora2 Strength Clip type: number x-order: 12 example: 0 lora2_strength_model: default: 2 description: LoRA 2 (add-detail-xl) model strength maximum: 2 minimum: 0 title: Lora2 Strength Model type: number x-order: 11 example: 0 lora3_strength_clip: default: 1 description: LoRA 3 (detailed_notrigger) clip strength maximum: 2 minimum: 0 title: Lora3 Strength Clip type: number x-order: 14 example: 0 lora3_strength_model: default: 1 description: LoRA 3 (detailed_notrigger) model strength maximum: 2 minimum: 0 title: Lora3 Strength Model type: number x-order: 13 example: 0 negative_prompt: default: ugly, deformed, soft, ext, watermark, abstract, big hands, fake, fake hands, distorted, drawing, painting, crayon, sketch, impressionist, worst quality, low quality, normal quality, lowres, low details, oversaturated, undersaturated, overexposed description: Things you do not want to see in your image title: Negative Prompt type: string x-order: 2 example: ugly, deformed, soft, ext, watermark, abstract, big hands, fake, fake hands, distorted, drawing, painting, crayon, sketch, impressionist, worst quality, low quality, normal quality, lowres, low details, oversaturated, undersaturated, overexposed output_format: default: webp description: Format of the output images enum: - webp - jpg - png title: output_format type: string x-order: 24 example: webp output_quality: default: 95 description: Quality of the output images, from 0 to 100. 100 is best quality, 0 is lowest quality. maximum: 100 minimum: 0 title: Output Quality type: integer x-order: 25 example: 95 prompt: default: high quality, detailed, photorealistic, perfect eyes, bokeh, photo, perfect light, detailed style, depth of field description: High quality modifiers to add to prompt title: Prompt type: string x-order: 1 example: squirrel sitting on a table with a candle and a bottle of perfume, highly detailed scene, highly detailed composition, high quality, detailed, photorealistic, perfect eyes, bokeh, photo, perfect light, detailed style, depth of field safety_filter: default: true description: Enable Replicate safety filter to detect and filter NSFW content (can't be disabled in web UI) title: Safety Filter type: boolean x-order: 21 example: true sampler_name: default: euler description: Sampling algorithm to use enum: - euler - euler_ancestral - heun - dpm_2 - dpm_2_ancestral - lms - dpm_fast - dpm_adaptive - dpmpp_2s_ancestral - dpmpp_sde - dpmpp_sde_gpu - dpmpp_2m - dpmpp_2m_sde - dpmpp_2m_sde_gpu - dpmpp_3m_sde - dpmpp_3m_sde_gpu - ddpm - ddim title: sampler_name type: string x-order: 7 example: ddim scheduler: default: karras description: Scheduler for the sampler enum: - normal - karras - exponential - sgm_uniform - simple - ddim_uniform title: scheduler type: string x-order: 8 example: sgm_uniform seed: description: Set a seed for reproducibility. Random by default. title: Seed type: integer x-order: 26 example: 1 steps: default: 50 description: Number of inference steps maximum: 100 minimum: 1 title: Steps type: integer x-order: 4 example: 50 width: default: 1024 description: Width of the image maximum: 2048 minimum: 512 title: Width type: integer x-order: 22 example: 2048 title: Input type: object Output: items: format: uri type: string title: Output type: array PredictionRequest: properties: created_at: format: date-time title: Created At type: string id: title: Id type: string input: properties: cfg: default: 7.5 description: Guidance scale maximum: 20 minimum: 1 title: Cfg type: number x-order: 5 checkpoint_name: default: realismEngineSDXL_v30VAE.safetensors description: Checkpoint model to use enum: - cyberrealisticXL_v5.safetensors - albedobaseXL_v31Large.safetensors - realDream_sdxl5.safetensors - realismEngineSDXL_v30VAE.safetensors - realvisxlV50_v50Bakedvae.safetensors title: checkpoint_name type: string x-order: 3 controlnet_end_percent: default: 0.5 description: ControlNet end percentage maximum: 1 minimum: 0 title: Controlnet End Percent type: number x-order: 20 controlnet_start_percent: default: 0 description: ControlNet start percentage maximum: 1 minimum: 0 title: Controlnet Start Percent type: number x-order: 19 controlnet_strength: default: 1 description: ControlNet strength maximum: 2 minimum: 0 title: Controlnet Strength type: number x-order: 18 denoise: default: 0.8 description: Denoising strength maximum: 1 minimum: 0 title: Denoise type: number x-order: 6 height: default: 1024 description: Height of the image maximum: 2048 minimum: 512 title: Height type: integer x-order: 23 input_image: description: Input image for processing. If not provided, the default image will be used. format: uri title: Input Image type: string x-order: 0 ipadapter_end_at: default: 1 description: IP-Adapter end at step percentage maximum: 1 minimum: 0 title: Ipadapter End At type: number x-order: 17 ipadapter_start_at: default: 0 description: IP-Adapter start at step percentage maximum: 1 minimum: 0 title: Ipadapter Start At type: number x-order: 16 ipadapter_weight: default: 0.9 description: IP-Adapter weight maximum: 1 minimum: 0 title: Ipadapter Weight type: number x-order: 15 lora1_strength_clip: default: 1 description: LoRA 1 (product marketing) clip strength maximum: 2 minimum: 0 title: Lora1 Strength Clip type: number x-order: 10 lora1_strength_model: default: 1 description: LoRA 1 (product marketing) model strength maximum: 2 minimum: 0 title: Lora1 Strength Model type: number x-order: 9 lora2_strength_clip: default: 1 description: LoRA 2 (add-detail-xl) clip strength maximum: 2 minimum: 0 title: Lora2 Strength Clip type: number x-order: 12 lora2_strength_model: default: 2 description: LoRA 2 (add-detail-xl) model strength maximum: 2 minimum: 0 title: Lora2 Strength Model type: number x-order: 11 lora3_strength_clip: default: 1 description: LoRA 3 (detailed_notrigger) clip strength maximum: 2 minimum: 0 title: Lora3 Strength Clip type: number x-order: 14 lora3_strength_model: default: 1 description: LoRA 3 (detailed_notrigger) model strength maximum: 2 minimum: 0 title: Lora3 Strength Model type: number x-order: 13 negative_prompt: default: ugly, deformed, soft, ext, watermark, abstract, big hands, fake, fake hands, distorted, drawing, painting, crayon, sketch, impressionist, worst quality, low quality, normal quality, lowres, low details, oversaturated, undersaturated, overexposed description: Things you do not want to see in your image title: Negative Prompt type: string x-order: 2 output_format: default: webp description: Format of the output images enum: - webp - jpg - png title: output_format type: string x-order: 24 output_quality: default: 95 description: Quality of the output images, from 0 to 100. 100 is best quality, 0 is lowest quality. maximum: 100 minimum: 0 title: Output Quality type: integer x-order: 25 prompt: default: high quality, detailed, photorealistic, perfect eyes, bokeh, photo, perfect light, detailed style, depth of field description: High quality modifiers to add to prompt title: Prompt type: string x-order: 1 safety_filter: default: true description: Enable Replicate safety filter to detect and filter NSFW content (can't be disabled in web UI) title: Safety Filter type: boolean x-order: 21 sampler_name: default: euler description: Sampling algorithm to use enum: - euler - euler_ancestral - heun - dpm_2 - dpm_2_ancestral - lms - dpm_fast - dpm_adaptive - dpmpp_2s_ancestral - dpmpp_sde - dpmpp_sde_gpu - dpmpp_2m - dpmpp_2m_sde - dpmpp_2m_sde_gpu - dpmpp_3m_sde - dpmpp_3m_sde_gpu - ddpm - ddim title: sampler_name type: string x-order: 7 scheduler: default: karras description: Scheduler for the sampler enum: - normal - karras - exponential - sgm_uniform - simple - ddim_uniform title: scheduler type: string x-order: 8 seed: description: Set a seed for reproducibility. Random by default. title: Seed type: integer x-order: 26 steps: default: 50 description: Number of inference steps maximum: 100 minimum: 1 title: Steps type: integer x-order: 4 width: default: 1024 description: Width of the image maximum: 2048 minimum: 512 title: Width type: integer x-order: 22 title: Input type: object output_file_prefix: title: Output File Prefix type: string webhook: format: uri maxLength: 65536 minLength: 1 title: Webhook type: string webhook_events_filter: default: - start - output - logs - completed items: description: An enumeration. enum: - start - output - logs - completed title: WebhookEvent type: string type: array title: PredictionRequest type: object PredictionResponse: properties: completed_at: format: date-time title: Completed At type: string created_at: format: date-time title: Created At type: string error: title: Error type: string id: title: Id type: string input: properties: cfg: default: 7.5 description: Guidance scale maximum: 20 minimum: 1 title: Cfg type: number x-order: 5 checkpoint_name: default: realismEngineSDXL_v30VAE.safetensors description: Checkpoint model to use enum: - cyberrealisticXL_v5.safetensors - albedobaseXL_v31Large.safetensors - realDream_sdxl5.safetensors - realismEngineSDXL_v30VAE.safetensors - realvisxlV50_v50Bakedvae.safetensors title: checkpoint_name type: string x-order: 3 controlnet_end_percent: default: 0.5 description: ControlNet end percentage maximum: 1 minimum: 0 title: Controlnet End Percent type: number x-order: 20 controlnet_start_percent: default: 0 description: ControlNet start percentage maximum: 1 minimum: 0 title: Controlnet Start Percent type: number x-order: 19 controlnet_strength: default: 1 description: ControlNet strength maximum: 2 minimum: 0 title: Controlnet Strength type: number x-order: 18 denoise: default: 0.8 description: Denoising strength maximum: 1 minimum: 0 title: Denoise type: number x-order: 6 height: default: 1024 description: Height of the image maximum: 2048 minimum: 512 title: Height type: integer x-order: 23 input_image: description: Input image for processing. If not provided, the default image will be used. format: uri title: Input Image type: string x-order: 0 ipadapter_end_at: default: 1 description: IP-Adapter end at step percentage maximum: 1 minimum: 0 title: Ipadapter End At type: number x-order: 17 ipadapter_start_at: default: 0 description: IP-Adapter start at step percentage maximum: 1 minimum: 0 title: Ipadapter Start At type: number x-order: 16 ipadapter_weight: default: 0.9 description: IP-Adapter weight maximum: 1 minimum: 0 title: Ipadapter Weight type: number x-order: 15 lora1_strength_clip: default: 1 description: LoRA 1 (product marketing) clip strength maximum: 2 minimum: 0 title: Lora1 Strength Clip type: number x-order: 10 lora1_strength_model: default: 1 description: LoRA 1 (product marketing) model strength maximum: 2 minimum: 0 title: Lora1 Strength Model type: number x-order: 9 lora2_strength_clip: default: 1 description: LoRA 2 (add-detail-xl) clip strength maximum: 2 minimum: 0 title: Lora2 Strength Clip type: number x-order: 12 lora2_strength_model: default: 2 description: LoRA 2 (add-detail-xl) model strength maximum: 2 minimum: 0 title: Lora2 Strength Model type: number x-order: 11 lora3_strength_clip: default: 1 description: LoRA 3 (detailed_notrigger) clip strength maximum: 2 minimum: 0 title: Lora3 Strength Clip type: number x-order: 14 lora3_strength_model: default: 1 description: LoRA 3 (detailed_notrigger) model strength maximum: 2 minimum: 0 title: Lora3 Strength Model type: number x-order: 13 negative_prompt: default: ugly, deformed, soft, ext, watermark, abstract, big hands, fake, fake hands, distorted, drawing, painting, crayon, sketch, impressionist, worst quality, low quality, normal quality, lowres, low details, oversaturated, undersaturated, overexposed description: Things you do not want to see in your image title: Negative Prompt type: string x-order: 2 output_format: default: webp description: Format of the output images enum: - webp - jpg - png title: output_format type: string x-order: 24 output_quality: default: 95 description: Quality of the output images, from 0 to 100. 100 is best quality, 0 is lowest quality. maximum: 100 minimum: 0 title: Output Quality type: integer x-order: 25 prompt: default: high quality, detailed, photorealistic, perfect eyes, bokeh, photo, perfect light, detailed style, depth of field description: High quality modifiers to add to prompt title: Prompt type: string x-order: 1 safety_filter: default: true description: Enable Replicate safety filter to detect and filter NSFW content (can't be disabled in web UI) title: Safety Filter type: boolean x-order: 21 sampler_name: default: euler description: Sampling algorithm to use enum: - euler - euler_ancestral - heun - dpm_2 - dpm_2_ancestral - lms - dpm_fast - dpm_adaptive - dpmpp_2s_ancestral - dpmpp_sde - dpmpp_sde_gpu - dpmpp_2m - dpmpp_2m_sde - dpmpp_2m_sde_gpu - dpmpp_3m_sde - dpmpp_3m_sde_gpu - ddpm - ddim title: sampler_name type: string x-order: 7 scheduler: default: karras description: Scheduler for the sampler enum: - normal - karras - exponential - sgm_uniform - simple - ddim_uniform title: scheduler type: string x-order: 8 seed: description: Set a seed for reproducibility. Random by default. title: Seed type: integer x-order: 26 steps: default: 50 description: Number of inference steps maximum: 100 minimum: 1 title: Steps type: integer x-order: 4 width: default: 1024 description: Width of the image maximum: 2048 minimum: 512 title: Width type: integer x-order: 22 title: Input type: object logs: default: "" title: Logs type: string metrics: title: Metrics type: object output: items: format: uri type: string title: Output type: array started_at: format: date-time title: Started At type: string status: description: An enumeration. enum: - starting - processing - succeeded - canceled - failed title: Status type: string version: title: Version type: string title: PredictionResponse type: object Status: description: An enumeration. enum: - starting - processing - succeeded - canceled - failed title: Status type: string ValidationError: properties: loc: items: anyOf: - type: string - type: integer title: Location type: array msg: title: Message type: string type: title: Error Type type: string required: - loc - msg - type title: ValidationError type: object WebhookEvent: description: An enumeration. enum: - start - output - logs - completed title: WebhookEvent type: string checkpoint_name: description: An enumeration. enum: - cyberrealisticXL_v5.safetensors - albedobaseXL_v31Large.safetensors - realDream_sdxl5.safetensors - realismEngineSDXL_v30VAE.safetensors - realvisxlV50_v50Bakedvae.safetensors title: checkpoint_name type: string output_format: description: An enumeration. enum: - webp - jpg - png title: output_format type: string sampler_name: description: An enumeration. enum: - euler - euler_ancestral - heun - dpm_2 - dpm_2_ancestral - lms - dpm_fast - dpm_adaptive - dpmpp_2s_ancestral - dpmpp_sde - dpmpp_sde_gpu - dpmpp_2m - dpmpp_2m_sde - dpmpp_2m_sde_gpu - dpmpp_3m_sde - dpmpp_3m_sde_gpu - ddpm - ddim title: sampler_name type: string scheduler: description: An enumeration. enum: - normal - karras - exponential - sgm_uniform - simple - ddim_uniform title: scheduler type: string securitySchemes: {} security: {} paths: /predictions: post: summary: Create a prediction with goodguy1963/sdxl-finetunes-img2img description: Run inference with the goodguy1963/sdxl-finetunes-img2img model on API.market operationId: createSdxl_finetunes_img2imgPrediction requestBody: required: true content: application/json: schema: type: object required: - version - input properties: version: type: string description: Model version ID default: 08807c2a252379c60ca5bc07e25d8f26fecfa9ad1d899987be8865e6d6a47ef2 input: $ref: "#/components/schemas/Input" examples: default: summary: Default example value: version: 08807c2a252379c60ca5bc07e25d8f26fecfa9ad1d899987be8865e6d6a47ef2 input: cfg: 2.47 steps: 50 width: 2048 height: 2048 prompt: squirrel sitting on a table with a candle and a bottle of perfume, highly detailed scene, highly detailed composition, high quality, detailed, photorealistic, perfect eyes, bokeh, photo, perfect light, detailed style, depth of field denoise: 0.8 scheduler: sgm_uniform input_image: https://replicate.delivery/pbxt/MkFZUXh7QUtpvY81ENJOEH0tueIRKIyn0jETUY8Q15RO29xe/DALL%C2%B7E%202023-10-27%2017.55.26%20-%20Lifestyle%20product%20photo%20showcasing%20a%20glistening%20perfume%20bottle%20on%20a%20wooden%20table%2C%20catching%20the%20ambient%20light.%20In%20the%20soft-focus%20background%2C%20a%20Swedish%20.png sampler_name: ddim output_format: webp safety_filter: true output_quality: 95 checkpoint_name: realvisxlV50_v50Bakedvae.safetensors negative_prompt: ugly, deformed, soft, ext, watermark, abstract, big hands, fake, fake hands, distorted, drawing, painting, crayon, sketch, impressionist, worst quality, low quality, normal quality, lowres, low details, oversaturated, undersaturated, overexposed ipadapter_end_at: 1 ipadapter_weight: 0.9 ipadapter_start_at: 0 controlnet_strength: 1 lora1_strength_clip: 1 lora2_strength_clip: 0 lora3_strength_clip: 0 lora1_strength_model: 0.5 lora2_strength_model: 0 lora3_strength_model: 0 controlnet_end_percent: 0.5 controlnet_start_percent: 0 responses: "201": description: Prediction created successfully content: application/json: schema: type: object properties: id: type: string description: Unique identifier for the prediction version: type: string description: Model version used for the prediction status: type: string enum: - starting - processing - succeeded - failed - canceled description: Current status of the prediction created_at: type: string format: date-time description: When the prediction was created examples: success: summary: Successful prediction creation value: id: ufawqhfynnddngldkgtslldrkq version: 08807c2a252379c60ca5bc07e25d8f26fecfa9ad1d899987be8865e6d6a47ef2 status: starting created_at: 2025-05-18T10:44:52.377Z "400": description: Bad request, validation error "401": description: Unauthorized, invalid API token parameters: - &a1 description: API.market API Key in: header name: x-api-market-key value: Please Login/Signup to get an API Key required: true schema: type: string /predictions/{id}: get: summary: Get prediction status and results description: Get the status and results of a goodguy1963/sdxl-finetunes-img2img prediction from API.market operationId: getSdxl_finetunes_img2imgPrediction parameters: - name: id in: path required: true schema: type: string description: The prediction ID responses: "200": description: Prediction details content: application/json: schema: type: object properties: id: type: string description: Unique identifier for the prediction status: type: string enum: - starting - processing - succeeded - failed - canceled description: Current status of the prediction output: $ref: "#/components/schemas/Output" error: type: string description: Error message if the prediction failed created_at: type: string format: date-time description: When the prediction was created started_at: type: string format: date-time description: When the prediction started processing completed_at: type: string format: date-time description: When the prediction completed examples: succeeded: summary: Successful prediction result value: id: ufawqhfynnddngldkgtslldrkq status: succeeded output: - https://replicate.delivery/xezq/fEGUWx6rBj1iGin9NY6qsMHEk05COT1epdAuwr6JrnRODe6oA/ComfyUI_00001_.webp created_at: 2023-01-01T12:00:00.000Z started_at: 2023-01-01T12:00:05.000Z completed_at: 2023-01-01T12:01:00.000Z "401": description: Unauthorized, invalid API token "404": description: Prediction not found parameters: - *a1 tags: - name: artificial-intelligence description: Operations related to artificial-intelligence - name: image description: Operations related to image - name: visual-recognition description: Operations related to visual-recognition - name: image-to-image description: Operations related to image-to-image