Flex.1-alpha-Redux Open-source Image Generation Model - Generate more detailed images, and it's free to deploy and use immediately!

Flex.1 Alpha Redux

Developed by ostris

Redux adapter trained on Flex.1-alpha, utilizing a 512x512 resolution SigLIP2 visual encoder for finer image generation

Image Generation Open Source License:Apache-2.0 #512-resolution visual encoding #SigLIP2 enhancement #Business-friendly license

Downloads 6,402

Release Time : 4/4/2025

Model Overview

This is an adapter model compatible with Flex.1-alpha and FLUX.1-dev, offering enhanced image generation capabilities

Model Features

Enhanced visual encoder

Uses a 512x512 resolution siglip2-so400m-patch16-512 visual encoder, capturing more details compared to the original 384 resolution

Business-friendly license

Adopts Apache 2.0 license, allowing unrestricted commercial use

Hardware-friendly

Supports fine-tuning and training on consumer-grade GPUs (e.g., 3090)

Model Capabilities

High-resolution image generation

Text-to-image conversion

Image detail enhancement

Use Cases

Creative design

Concept art creation

Generate high-detail concept art for games or films

512x512 high-resolution output

Commercial applications

Advertising material generation

Quickly produce high-quality images for marketing purposes

Business-friendly license

🚀 Flex.1-alpha 512 Redux Adapter

This is a Redux adapter trained from scratch on Flex.1-alpha, offering better image quality and an Apache 2.0 license for commercial use.

🚀 Quick Start

This adapter is a Redux adapter trained from scratch on Flex.1-alpha. It can also work with FLUX.1-dev.

Here is a visual of the adapter: Flex.1 Redux Comfy

✨ Features

Better Vision Encoder

The original Redux by BFL uses siglip-so400m-patch14-384. This adapter uses siglip2-so400m-patch16-512, which is newer and has an increased resolution of 512x512. This allows for better image quality and more detail in the generated images.

Apache 2.0 License

Since this vision encoder was trained from scratch on Flex.1-alpha, it can be licensed under the Apache 2.0 license. This means that you can use it for commercial purposes without any restrictions.

🔧 Technical Details

The Vision Encoder

Flex Redux uses the latest SigLIP2 vision encoder at a resolution of 512x512. This allows the vision encoder to see more fine details in the images.

Here is a comparison of the clip sizes: Compare blip sizes

📦 Installation

ComfyUI

You will need this custom node: ComfyUI-Advanced-Vision and the siglip2 vision encoder here: siglip2_so400m_patch16_512.safetensors. Put that inside your ComfyUI/models/clip_vision/ folder.

If you don't have Flex.1-alpha, you can pick it up here: Flex.1-alpha.safetensors. Put that in ComfyUI/models/checkpoints/. You can also use FLUX.1-dev if you prefer.

Put the flex1_redux_siglip2_512.safetensors file in this repo in ComfyUI/models/style_models/. There is also a workflow named flex-redux-workflow.json to get you started.

📚 Documentation

Fine tune it!

This adapter was 100% trained with AI Toolkit on a single 3090 over the course of a few weeks. Which means you can fine tune it, or train your own, on consumer hardware. Check out the example config to get started.