What It Does
EmptyLatentImage creates the starting latent for text-to-image workflows. The width and height parameters define the generation canvas in latent pixels (multiply by 8 for actual image pixels). The batch_size parameter allows generating multiple images in a single pass.
Dimensions are significant provenance data: they affect composition, aspect ratio, and quality. Models are trained on specific resolutions (SD 1.5 at 512×512, SDXL at 1024×1024), and generating at other resolutions can cause artifacts or quality degradation.
Inputs
widthINTCanvas width in latent pixels (÷8 for image pixels).
heightINTCanvas height in latent pixels.
batch_sizeINToptionalNumber of images to generate.
Outputs
LATENTLATENTEmpty latent tensor of specified dimensions.
What Numonic Captures
- Width and height (latent and effective pixel dimensions)
- Batch size
Known Gaps
- Optimal resolution for the selected model — no validation recorded
Related Nodes
Capture ComfyUI metadata automatically
Numonic extracts workflow metadata from every ComfyUI generation — models, samplers, seeds, prompts, and custom nodes. Track provenance, maintain compliance, and never lose a workflow.