🐳iPhone

Nostalgia is now a ComfyUI node. 🐳iPhone (iPhone Filter) ships 40+ color presets named after iPhone camera generations - Vivid, Dramatic, Mono, Noir, Silvertone from the iPhone 6 era, Natural and Studio Light from the iPhone 8/X era, the Vibrant/Warm/Cool looks of the 11–14, and the Radiant/Peaceful/Amber/Gold/Rose names of the 15 and 16 Pros. It's a LUT drawer, basically, without you having to hunt down and install actual LUT files.

Where it earns its keep: you want a consistent "shot on a phone camera" look across a batch - mockups for an app, social content, a set of product shots - and you want it applied identically to every frame. Pick a preset, and every image in the batch gets graded the same way. For that use it's genuinely nice.

How it works

Each preset is just a bundle of adjustment parameters - brightness, contrast, saturation, warmth, highlights, shadows, vignette, grain - that the node applies through a Pillow pipeline. Warmth shifts the red/blue balance, highlights and shadows do a tonal lift/crush, vignette darkens the corners radially, and grain adds a bit of noise. Presets are opinionated combinations of those numbers, tuned to mimic what each phone's default processing pushed toward.

If no preset does it for you, filter_preset offers "Custom," which unlocks the manual sliders instead:

• custom_brightness (0.5–2, default 1) and custom_contrast (0.5–2, default 1)
• custom_saturation (0–2, default 1)
• custom_warmth (-50 to +50)
• custom_highlights / custom_shadows (-100 to +100 each)
• custom_vignette (0–1) and custom_grain (0–50)

The one image output feeds anything downstream.

What to actually set

Honestly: pick a preset, run it, and if it's too strong, dial the relevant custom slider in the other direction from whatever the preset set. The presets are named and tuned, but they're not subtle - "Noir" is a hard black-and-white, "Dramatic" crushes shadows noticeably. Treat them as starting points, not final looks.

The honest take: this is a fun, shallow node. If you're doing real color work, a proper color-grading node or LUT loader gives you more control, and the phone "look" is mostly just a strong contrast+saturation curve plus a bit of grain. Where this wins is speed and instant recognizability for social-style content. It's also entirely CPU-side Pillow math - no model, no download, nothing heavy.

Install

Part of the Pond Nodes pack:

ComfyUI Manager → search "Pond Nodes" (comfy_Pond_Nodes)

or:

cd ComfyUI/custom_nodes
git clone https://github.com/Pondowner857/comfy_Pond_Nodes
cd comfy_Pond_Nodes
pip install -r requirements.txt

Restart after. Only Pillow and numpy matter for this node. And the pack's standing warning: if your terminal starts spamming after install, the README blames a conflict with comfyui_HiDream-Sampler - check that before suspecting this filter.