Florence 2 large PromptGen v1.5 EU AI Act Compliance Profile

Florence-2-large-PromptGen v1.5

This is a major version upgrade for PromptGen. In this new version, two new caption instructions are added: <GENERATE_TAGS> and <MIXED_CAPTION> You'll also notice significantly improved accuracy with this version, thanks to a new training dataset. It no longer relies on Civitai Data, avoiding the issues of lora trigger words and inaccurate tags from misinterpretation.

About PromptGen

Florence-2-large-PromptGen v1.5 is a model trained for MiaoshouAI Tagger for ComfyUI. It is an advanced image captioning tool based on the Microsoft Florence-2 Model Large and fine-tuned to perfection.

Why another tagging model?

Most vision models today are trained mainly for general vision recognition purposes, but when doing prompting and image tagging for model training, the format and details of the captions is quite different. Florence-2-large-PromptGen is trained on such a purpose as aiming to improve the tagging experience and accuracy of the prompt and tagging job. The model is trained based on images and cleaned tags from Civitai so that the end result for tagging the images are the prompts you use to generate these images. You won't get annoying captions like "This is image is about a girl..." or

Features:

• Describes image in much detail when using <MORE_DETAILED_CAPTION> instruction. <img style="width:100%; hight:100%" src="https://msdn.miaoshouai.com/miaoshou/bo/2024-09-05_12-40-34.png" />
• When using <DETAILED_CAPTION> instruction, it creates a structured caption with infomation on subject's position and also reads the text from the image, which can be super useful when recreate a scene. <img style="width:100%; hight:100%" src="https://msdn.miaoshouai.com/miaoshou/bo/2024-09-05_13-07-54.png" />
• Memory efficient compare to other models! This is a really light weight caption model that allows you to use a

Your obligations as a Florence 2 large PromptGen v1.5 deployer

Track your progress

Sign up to track obligations, receive deadline reminders, and generate audit packages.

Tags