Instagram Filters-Vextra

The Apply Instagram Filter node (displayed as "Instagram Filters-Vextra") does what the name promises: drop a Clarendon, a Nashville, a Valencia, or any of the classic Instagram filters onto your generated image, right inside ComfyUI. The name is a slight lie in the good way - it doesn't call any API and needs no key. All twenty-six filters are applied locally by the pilgram Python library, which is the same open-source implementation behind those look-at-me filters from a decade ago.

It's one of the pack's "Vextra" ports: diontimmer's ComfyUI-Vextra-Nodes had import errors that the maintainer never got around to fixing, so Aegis72 folded the working nodes in here with credit. This one's a straight port, no extensions.

How it works

Pick a filter from the dropdown and every image in the batch gets that filter applied via PIL, then converted back to a tensor. The filter menu is the classic lineup - _1977, aden, brannan, brooklyn, clarendon, earlybird, gingham, hudson, inkwell, kelvin, lark, lofi, maven, mayfair, moon, nashville, perpetua, reyes, rise, slumber, stinson, toaster, valencia, walden, willow, xpro2. If you grew up editing phone photos you know half of these already: inkwell is the black-and-white, clarendon is the brightened contrast bump, nashville is the faded warm look.

Inputs and outputs

• images (IMAGE, required) - the image or batch to filter.
• instagram_filter (enum, 26 choices) - which filter to apply.
• Output: IMAGE.

When you'd use it

Stylization and nostalgia, mostly. These filters are heavy-handed 2010s looks - great for a uniform social-media feel across a batch, less useful if you want subtle grading. If you're generating a set of images meant to look like they all came from the same phone, running the same filter over the whole batch is a one-node way to get there.

Installing it

Standard for the pack, with one extra wrinkle:

cd ComfyUI/custom_nodes
git clone https://github.com/aegis72/aegisflow_utility_nodes

Restart ComfyUI, or install "AegisFlow Utility Nodes" via ComfyUI Manager. The wrinkle: pilgram isn't in a requirements.txt - the node auto-installs it with pip the first time it loads. That's convenient until it isn't: on a locked-down or offline environment, the auto-install can fail and the node errors at import. If that happens, pip install pilgram in your ComfyUI venv by hand and restart.

Where people get burned

Don't expect subtlety. These are full-strength Instagram looks with no intensity slider - there's no "50% nashville" option, so if you want a lighter touch you'll have to dial it down with a blend or brightness node downstream. And because they're batch-applied identically, they'll also happily flatten the tonal differences you spent the whole workflow building. Fun tool, just keep the effect strength in mind.