pull down to refresh

I've tried to get AI to make this image for me, and I'm curious if I'm not understanding how to use the prompts or if I'm expecting too much from this tool in the first place.

5000 sats go to the first person to generate this accurately.

Not posting as a bounty because I personally don't think it's possible, but I'll keep my word if someone is able to do it. Either way, interested in the discussion around this to better understand these tools.

Details: A warrior standing in the woods, overcast lighting Wearing a dark gray cloak and dark green armor, can’t see face Holding a bo staff, which has an emerald embedded in the center Wearing a belt with a few sheathed knives around waist Wearing a bow and quiver slung over back

this territory is moderated

fwiw, these are as close as I got. There are several details I just can't seem to get it to do: Dark gray cloak AND dark green clothes (will usually do one or the other) An actual bo staff, let alone with a gem embedded in the right place Knives on the belt https://m.stacker.news/19018 https://m.stacker.news/19019

reply

I used the tool below that is very good for several images

https://huggingface.co/spaces/black-forest-labs/flux.1-dev

reply

https://m.stacker.news/19058

I kinda like the lighting on this one. Tweaked the prompt a bit, but its hard to hit all those points.

Model: SDXL

reply

This one hits more of the points but his weapon stance is all messed up

https://m.stacker.news/19059

reply

cc @nym, I know you're good at this stuff

reply

Thanks for the heads up! I’ll give it a try!

reply