ComfyUI-RMAutomation

RMFaceComposite

Collects all independently processed faces (via INPUT_IS_LIST) and composites them sequentially onto the original image. Face 0 is composited first, then face 1 onto that result, and so on. Each face uses its own Gaussian-blurred mask for smooth edge blending. When face_data indicates no faces were found, passes the original image through unchanged. Replaces: Upscale Image + Gaussian Blur Mask + ImageCompositeMasked

RMAutomation/Face · 3 inputs · 2 parameters · 1 output

RMFaceDetectCrop

Detects faces in an image, crops each face region (expanded by crop_factor), and scales to max_size. Uses OUTPUT_IS_LIST for the image output so ComfyUI executes all downstream nodes once per face — each face gets its own VAE encode, guide, sampler pass, and VAE decode independently. Replaces: UltralyticsDetectorProvider + BBOX Detector (SEGS) + Decompose (SEGS) + From SEG_ELT + From SEG_ELT crop_region + Bounding Box + Simple Math x2 + CropByBBoxes + ImageScaleToMaxDimension

RMAutomation/Face · 3 outputs

RMImageFallback

Returns the first non-empty image from multiple inputs. Useful for conditional workflows.

RMAutomation/Image · 3 inputs · 0 parameters · 1 output

RMLoraApply

Applies all LoRAs from a collected stack to the model and CLIP.

RMAutomation/LoRA · 1 input · 0 parameters · 2 outputs

RMLoraCollector

Combines multiple LoRA stacks into one. Accepts both LORASTACK inputs and lora strings.

RMAutomation/LoRA · 8 inputs · 0 parameters · 1 output

RMMakeImageBatch

Combine any number of images into a single image batch.

RMAutomation/Image · 0 inputs · 0 parameters · 1 output

RMMaskCombine

Combines body, face, and dynamic object masks into a single mask. Use "Add Object Input" to add mask inputs for body parts to exclude. White = process, Black = protect. The result masks the background + face, excluding the body and objects.

RMAutomation/Mask · 0 inputs · 0 parameters · 1 output

RMMaskDenoiseRemap

Remaps a mask so that black and white regions map to custom denoise levels. Feed the output into Set Latent Noise Mask for per-region denoise control. Output = black_denoise + mask * (white_denoise - black_denoise)

RMAutomation/Mask · 1 input · 2 parameters · 1 output

RMMaskGate

Validates masks and passes through only valid ones. Outputs None for invalid masks, preventing crashes in downstream nodes.

RMAutomation/Mask · 1 input · 0 parameters · 1 output

RMMaskGateGuide

Same as Mask Gate but for guide inputs. Disables guides when masks are invalid.

RMAutomation/Mask · 2 inputs · 0 parameters · 1 output

RMMathExpression

Compact description unavailable for this node.

RMAutomation/Utils · 2 outputs

RMMultiDetectCrop

Detects objects using multiple BBOX models (each with its own threshold), crops each detection region (expanded by crop_factor), and scales to max_size. Uses OUTPUT_IS_LIST so each detection flows independently through downstream nodes.

RMAutomation/Detection · 1 input · 4 parameters · 3 outputs

RMNegativeTextEmbed

Combines an optional string input with a text field and encodes to negative conditioning.

RMAutomation/Text · 2 inputs · 1 parameter · 1 output

RMPositiveTextEmbed

Combines an optional string input with a text field and encodes to positive conditioning.

RMAutomation/Text · 2 inputs · 1 parameter · 1 output

RMPowerLoraLoader

A flexible node for loading multiple LoRAs with support for: - Dynamic number of LoRA inputs (add as many as you need) - Fixed or random strength per LoRA - Style string import for automated LoRA loading

RMAutomation/LoRA · 0 inputs · 0 parameters · 1 output

RMPowerLoraLoaderPipe

Power LoRA Loader variant that accepts a Style Pipe input. Allows selecting which lora type to use from the pipe.

RMAutomation/LoRA · 0 inputs · 0 parameters

RMSaveImageWithMetadata

Save images with metadata using RM_STYLES_PIPE. RM_STYLES_PIPE order: Positive, Negative, Motion, Image Loras, Motion High, Motion Low, Prompt Number, Prompt Name

RMAutomation/Image · 2 inputs · 2 parameters · 0 outputs

RMSetLatentNoiseMask

Improved latent noise mask setter with proper error handling for edge cases.

RMAutomation/Latent · 2 inputs · 0 parameters · 2 outputs

RMStylesFull

Load prompts from a JSON styles file with multiple selection modes: - Manual: Use specific prompt number - Random: Random selection within range - Increment: Auto-increment after each generation - Decrement: Auto-decrement after each generation Features an Edit Styles button that opens a modal editor for managing your styles.

RMAutomation/Styles · 0 inputs · 7 parameters · 8 outputs

RMStylesFullDisplay

Same as RM Styles Full but displays the current prompt on the node.

0 inputs · 7 parameters

RMStylesPipe

Same as RMStylesFull but outputs a pipe tuple for easy data passing. The pipe contains all style data in a single output (without PDXL prefixes).

0 inputs · 6 parameters · 1 output

RMStylesPipeOut

Unpack an RM_STYLES_PIPE into individual outputs. Optionally apply PDXL (Pony) quality tags to positive/negative prompts.

RMAutomation/Styles · 1 input · 1 parameter · 8 outputs

RMVideoCombine

Combines images into video with: - Multiple format support (H.264, H.265, VP9, GIF, WebP) - Configurable output directory - Metadata embedding - Audio support - Ping-pong looping

RMAutomation/Video · 1 output