littletry/nsfw-filter-custom

An adjustable image NSFW filter based on the Stable Diffusion v1.5 safety checker. Supports a configurable threshold offset while preserving the original special-care behavior.

Public
134.3K runs
Playground API Examples README Versions

Run time and cost

This model costs approximately $0.00022 to run on Replicate, or 4545 runs per $1, but this varies depending on your inputs. It is also open source and you can run it on your own computer with Docker.

This model runs on Nvidia T4 GPU hardware. Predictions typically complete within 1 seconds.

Readme

Attribution and licenses

This is an unofficial modified fork and is not affiliated with or endorsed by the original authors.

This model is based on:

The original Cog model code is licensed under the MIT License.

The Stable Diffusion v1.5 safety-checker weights are licensed under the CreativeML OpenRAIL-M License. Users must comply with its use-based restrictions.

This version adds an adjustable threshold_offset input. 0.0 matches the original threshold, while higher values make normal NSFW detection less strict. Special-care detection is not relaxed by default. Set strict_special_care=false to keep returning special-care labels without letting them strengthen normal NSFW concept matching.

More AI tools

Explore leading AI models online with SeeAPI — a unified platform for multimodal generation.

Model created
Model updated