Version 3.2 was trained with JuggernautXL v6 as a base, using a dataset of around 10k images, selected to teach concepts that were missing from the model. The main focuses of this model are nsfw, holding weapons, and horror/monster themes, (which is where the "unsafe" name comes from) but it is meant to be a generalist model capable of handling images with or without those concepts. Note that this version has an SDXL Turbo + LCM alternative, which can generate much faster at a cost to quality. A recommended workflow is to generate using the turbo version until you find a base image you like, then upscale it with the normal version.
The data sets I used were tagged using Danbooru tags, so that is the tagging style I used with the generated images, but CLIP-style tagging should still work. The model was also trained at various aspect ratios at 256, 512, 1024, and 2048 pixels, so it is capable of outputting at all those resolutions, but it obviously will have more room to play with at higher ones.