一个受Nano Banana的手办生成风格启发的kontext模型，使用“Convert the objects in the image to the nano_banana figure style”来将图片转换为Nano Banana风格的手办。
Strength: 1-1.2
Guidance: 3.6-4.0
Description
FAQ
Comments (10)
that's explain lot of fake anime figure images on internet. super cool, one for illustrious
+fake Blender 3D viewport on the monitor
@2P2 I think this reflects that the base capabilities of FLUX kontext are quite good. There are already many LoRAs with this anime figure style, but it seems only kontext can imitate nano banana's fake Blender 3D viewport. I've seen some Qwen image editing models that can't achieve this effect yet.
can you share the work flow please, my results dont come close to yours
I don't know how to upload the workflow, so I uploaded a picture with the workflow at the end of the album. The training set I used uses real-life pictures, so it may not work well on anime pictures. Please try to use pictures with the whole body or more than 3/4 of the body in the frame. If the effect is not good, you can try to increase the Strength value.
"Convert the objects in the image to the nano_banana figure style. isolate the subject. remove the background and make his like a small action figure statue on a desk near a computer"
this worked for me using a medium shot
@LatentDream That's good
@norsizu thank you :)
Results aren't clearly much better with the lora compared to just prompting for this kind of images sadly. But thanks for making it anyways
Yes, I mainly want to verify the basic capabilities of Flux kontext
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.