Skip to main content
Glama

post-img2img-ip-adapter-inferences

Generate new images by combining an initial image with reference images using IP Adapter models for style or character adaptation.

Instructions

Trigger a new image generation in Img2Img + IpAdapter mode. The first image is used to initialize the generation, and the second reference image is used as an IPAdapter.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
originalAssetsNoIf set to true, returns the original asset without transformation
dryRunNo
ipAdapterImageIdsNo
seedNoUsed to reproduce previous results. Default: randomly generated number.
strengthNoControls the noise intensity introduced to the input image, where a value of 1.0 completely erases the original image's details. Available for img2img and inpainting. (within [0.01, 1.0], default: 0.75)
modelIdYesThe model id to use for the inference
ipAdapterTypeNoThe type of IP Adapter model to use. Must be one of [`style`, `character`], default to `style``
ipAdapterImageNoDeprecated for type txt2img-ip-adapter and img2img-ip-adapter, use `ipAdapterImages` instead. The IpAdapter image as a data url. Will be ignored if the `ipAdapterImages` parameter is provided.
schedulerNoThe scheduler to use to override the default configured for the model. See detailed documentation for more details.
ipAdapterImagesNo
imageParentIdNoSpecifies the parent asset Id for the image when provided as a dataurl.
baseModelIdNoThe base model to use for the inference. Only Flux LoRA models can use this parameter. Allowed values are available in the model's attribute: `compliantModelIds`
heightNoThe height of the generated images, must be a 8 multiple (within [64, 2048], default: 512) If model.type is `sd-xl`, `sd-xl-lora`, `sd-xl-composition` the height must be within [512, 2048] If model.type is `sd-1_5`, the height must be within [64, 1024] If model.type is `flux.1.1-pro-ultra`, you can use the aspectRatio parameter instead
imageHideNoToggles the hidden status of the image when provided as a dataurl.
maskNoThe mask as a data URL, used to determine the area of change. The mask is a binary mask made out of white and black pixels. The white area is the one that will be replaced. (example: "data:image/png;base64,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")
imageNoThe input image as a data URL (example: "data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAAEAAAABCAQAAAC1HAwCAAAAC0lEQVQYV2NgYAAAAAMAAWgmWQ0AAAAASUVORK5CYII=") or the asset ID (example: "asset_GTrL3mq4SXWyMxkOHRxlpw")
imageIdNoDeprecated: The input image as an AssetId. Prefer to use image with the asset ID instead.
ipAdapterImageIdNoDeprecated for type txt2img-ip-adapter and img2img-ip-adapter, use `ipAdapterImageIds` instead. The IpAdapter image as an AssetId. Cannot be set if `ipAdapterImage` is provided. Will be ignored if the `ipAdapterImageIds` parameter is provided.
ipAdapterScaleNoDeprecated for type txt2img-ip-adapter and img2img-ip-adapter, use `ipAdapterScales` instead. IpAdapter scale factor (within [0.0, 1.0], default: 0.9). Will be ignored if the `ipAdapterScales` parameter is provided
modelEpochNoThe epoch of the model to use for the inference. Only available for Flux Lora Trained models.
hideResultsNoIf set, generated assets will be hidden and not returned in the list of images of the inference or when listing assets (default: false)
ipAdapterScalesNo
maskIdNoThe mask as an AssetId. Will be ignored if the `image` parameter is provided
negativePromptNoThe prompt not to guide the image generation, ignored when guidance < 1 (example: "((ugly face))") For Flux based model (not Fast-Flux): requires negativePromptStrength > 0 and active only for inference types txt2img / img2img / controlnet.
intermediateImagesNoEnable or disable the intermediate images generation (default: false)
conceptsNo
guidanceNoControls how closely the generated image follows the prompt. Higher values result in stronger adherence to the prompt. Default and allowed values depend on the model type: - For Flux dev models, the default is 3.5 and allowed values are within [0, 10] - For Flux pro models, the default is 3 and allowed values are within [2, 5] - For SDXL models, the default is 6 and allowed values are within [0, 20] - For SD1.5 models, the default is 7.5 and allowed values are within [0, 20]
numInferenceStepsNoThe number of denoising steps for each image generation (within [1, 150], default: 30)
numSamplesNoThe number of images to generate (within [1, 128], default: 4)
widthNoThe width of the generated images, must be a 8 multiple (within [64, 2048], default: 512) If model.type is `sd-xl`, `sd-xl-lora`, `sd-xl-composition` the width must be within [512, 2048] If model.type is `sd-1_5`, the width must be within [64, 1024] If model.type is `flux.1.1-pro-ultra`, you can use the aspectRatio parameter instead
negativePromptStrengthNoOnly applicable for flux-dev based models for `txt2img`, `img2img`, and `controlnet` inference types. Controls the influence of the negative prompt. Default 0 means the negative prompt has no effect. Higher values increase negative prompt influence. Must be > 0 if negativePrompt is provided.
promptYesFull text prompt including the model placeholder. (example: "an illustration of phoenix in a fantasy world, flying over a mountain, 8k, bokeh effect")

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/pasie15/scenario.com-mcp-server'

If you have feedback or need assistance with the MCP directory API, please join our Discord server