Flex-Redux / README.md
shehzad1's picture
Upload folder using huggingface_hub
45e6cf5 verified
---
language:
- en
license: other
license_name: flux-1-dev-non-commercial-license
license_link: LICENSE.md
tags:
- image-generation
- flux
- siglip2
---
**Input Image** | **Original FLUX Redux** | **SigLIP2-Enhanced**
:--:|:--:|:--:
<img src="./assets/mona_lisa.png" width="200" height="200"/> | <img src="./assets/siglip_output.png" width="200" height="200"/> | <img src="./assets/siglip2_output.png" width="200" height="200"/>
# Flex.1-alpha-Redux [dev] — Enhanced with SigLIP-2
Flex.1-alpha-Redux [dev] is an adapter for all FLUX.1 base models for image variation generation.
Given an input image, Flex.1 Redux can reproduce the image with slight variation, allowing for refined visual generation. It integrates naturally into more complex workflows, enabling image restyling and editing via text.
This version of Redux has been updated with modules from [`SigLIP2`](https://huggingface.co/google/siglip2-so400m-patch16-512) replacing the following components:
- [`image_encoder`](https://huggingface.co/ostris/ComfyUI-Advanced-Vision/blob/main/clip_vision/siglip2_so400m_patch16_512.safetensors)
- [`feature_extractor`](https://huggingface.co/google/siglip2-so400m-patch16-512/blob/main/preprocessor_config.json)
- [`image_embedder`](https://huggingface.co/ostris/Flex.1-alpha-Redux/blob/main/flex1_redux_siglip2_512.safetensors)
---