Open film canisters on a steel table in crimson studio light

Open source

Horror
LoRAs

Preview weights from Darkframe AI — found footage and Italian giallo today, more subgenres in training.

Published models · training more

Preview · CreativeML OpenRAIL-M

DARKFRAME — Lost Footage

Night-vision cast, VHS glitches, grain, and claustrophobic “discovered recording” stills.

Trigger · darkframe_analog

Model card →

Preview · CreativeML OpenRAIL-M

DARKFRAME — Italian Horror Giallo

Giallo gel lighting, razor tension, and stylized Italian horror color.

Trigger · darkframe_giallo

Model card →

We publish horror-native checkpoints on huggingface.co/darkframeai. Follow the org for new LoRAs across body horror, folk horror, vampires, and more.

FAQ

Where are Darkframe open-source models?
On Hugging Face under the darkframeai organization: huggingface.co/darkframeai. Current preview LoRAs are Lost Footage and Italian Horror Giallo.
What license are the Darkframe LoRAs?
Both published preview weights use CreativeML OpenRAIL-M. Read the model card on Hugging Face before commercial use.
What trigger words do the LoRAs use?
Lost Footage uses darkframe_analog. Italian Horror Giallo uses darkframe_giallo. Scale 1.2–1.4 is recommended on Lost Footage.