Unified-Vision-Prompt-Generator

Unified-Vision-Prompt-Generator
★ 1

图像提示生成多模型支持LLM集成ComfyUI节点
基于多种视觉LLM(Qwen2/2.5、Moondream、SmolVLM)的ComfyUI提示/图注生成器,快速生成高质量图像描述与提示以提升生成效果。
💡 自动为图像生成描述性提示以提升图像生成质量
🍴 1 Forks💻 Python🔄 2025-11-20
📦
网盘下载
复制链接后前往夸克网盘下载
https://pan.quark.cn/s/79aaff81621b
📄 README

A prompt generator for ComfyUI using various different LLM models

‘qwen2vl’: [

‘Qwen/Qwen2-VL-2B-Instruct’,

‘Qwen/Qwen2-VL-7B-Instruct’,

‘Qwen/Qwen2.5-VL-3B-Instruct’,

‘Qwen/Qwen2.5-VL-7B-Instruct’,

‘Qwen/Qwen2.5-VL-72B-Instruct’,

],

‘moondream’: [

‘vikhyatk/moondream2’,

‘vikhyatk/moondream1’,

],

‘smolvlm’: [

‘HuggingFaceTB/SmolVLM-Instruct’,

‘HuggingFaceTB/SmolVLM-256M-Instruct’,

‘HuggingFaceTB/SmolVLM-500M-Instruct’,

I am a bit limited with the amount of VRAM (16GB) that I have, so unable to check any of the high parameter Qwen models, os if you have used them, let me know if they are doing what they should be