Usage in ComfyUI
What is the recommended way to use this model in ComfyUI?
I was able to use it by replacing a pre-downloaded Florence2 folder with this cloned folder and it works, although it seems the folder isn't detected/supported natively by the DownloadAndLoadFlorence2Model node?
Oh I see it shows up in the Florence2ModelLoader node... all goods
Not sure what your use case / task is, but I’d recommend some of my newer models. All of my models are trained with both SFW & NSFW datasets.
I’d recommend my florence-2-base-interleaved model for tags & captions. Use the following tasks for a variety of outputs.
<TAGS>
<NSFW_TAGS>
<NSFW_CAPTION>
<NSFW_DETAILED_CAPTION>
<DOCCI_CAPTION>
<GPT4V_CAPTION>
<HUMAN_CAPTION>
<LLAVA_CAPTION>
Not sure what your use case / task is, but I’d recommend some of my newer models. All of my models are trained with both SFW & NSFW datasets.
I’d recommend my florence-2-base-interleaved model for tags & captions. Use the following tasks for a variety of outputs.
<TAGS>
<NSFW_TAGS>
<NSFW_CAPTION>
<NSFW_DETAILED_CAPTION>
<DOCCI_CAPTION>
<GPT4V_CAPTION>
<HUMAN_CAPTION>
<LLAVA_CAPTION>
Hey, hope it's OK I ask a question here, since this thread is relevant to my question. I followed your advice and downloaded the florence-2-base-interleaved model and selected it in my "Florence2ModelLoader" node in ComfyUI, which seems to work fine. However, I am unable to select any of the tasks you listed here. Do I have to type them manually somewhere? As you can see on the screenshot, I do not get any of those options. Hope you can help me out. :)