Depth Styler

DepthStyler is a 20-entry menu of depth-of-field presets - Shallow, Deep, Macro, Extreme Shallow, Moderate, Landscape, Close-Up, Environmental Portrait, and a few more. In photographer terms it's the "how much of the scene is in focus" dial, expressed as prompt words. Want that creamy blurred background behind a portrait? Point it at a shallow-DoF entry. Want a landscape sharp front to back? Deep DoF. It's a focused little tool (pun accepted) for one specific look.

This is a genuinely photographic node, and depth-of-field words are the kind of concrete optical language that diffusion models tend to respond to well - better than vague quality tags, anyway.

How it works

MileHighStyler is a fork of SDXL Prompt Styler. Each depth-of-field preset is a JSON template with a {prompt} slot; the node drops your text_positive in and wraps the DoF phrase at a strong weight (~2.0). Because "shallow depth of field," "bokeh," and "macro" are real photographic terms the model has seen in countless captions, they steer reliably. Output is a rewritten prompt string into your CLIP Text Encode; nothing touches the model.

The inputs and outputs that matter

• depth - the 20-entry depth-of-field dropdown. No option is the off switch.
• text_positive - your subject; the DoF cue attaches to it.
• text_negative - optional negative text.

log_prompt prints the finished prompt to the console. Outputs are text_positive and text_negative strings, into a CLIP Text Encode.

Installing it

ComfyUI Manager: search ComfyUI_MileHighStyler, install, restart. Or:

cd ComfyUI/custom_nodes && git clone https://github.com/TripleHeadedMonkey/ComfyUI_MileHighStyler

then restart. No models, no requirements.txt, no dependencies - Python and JSON. Nodes appear under ali1234/stylers.

Where people get burned

• Shallow DoF eats detail. If you push "extreme shallow depth of field" hard, the model blurs everything that isn't the subject - sometimes including parts of the subject you wanted sharp. If your background is too melted, that's the emphasis; ease off.
• It competes with your camera cue. Stack this with a CameraStylerAdvanced body and a filter and you've got three optical cues arguing. Depth-of-field usually wins the background; just be aware they interact.
• Model era. The 2.0 weighting is SDXL/SD 1.5 syntax and gets silently discarded on 2026 LLM-encoded models (Flux 2, Z-Image, Qwen). The DoF phrase itself still works there as plain language - "shallow depth of field, blurred background" reads fine as an instruction.
• Depth of field is not a depth map. Despite the name, this has nothing to do with ControlNet depth or depth estimation - it's purely a focus-look prompt cue. Don't wire it expecting a depth image.