OSS Alternative - Discover Top Open Source Alternatives to Popular Software

Hunyuan-PromptEnhancer/PromptEnhancer

A prompt rewriting tool that refines user prompts into clearer, structured versions to enhance the quality of text-to-image generation and image-to-image editing.

Core Features

Dual-mode support for Text-to-Image and Image-to-Image prompt enhancement.

Preserves original intent while restructuring prompts for clarity.

Robust parsing with multi-level fallback mechanisms.

Flexible deployment supporting full-precision, quantized (GGUF), and vision-language models.

Quick Start

pip install -r requirements.txt

Detailed Introduction

PromptEnhancer is an innovative utility developed by Tencent Hunyuan, designed to significantly improve the output quality of text-to-image and image-to-image models. By employing a chain-of-thought prompt rewriting approach, it transforms ambiguous or simple user inputs into detailed, structured prompts. This process ensures that the core intent of the original prompt is maintained while providing clearer instructions to downstream generative AI models, leading to more accurate and higher-fidelity image generation and editing results. It supports various model deployments for broad applicability.