Version 2.0 update: after I connected the grid pattern sometimes visible in bright or dark parts of the images with the fp4 training with OneTrainer, I switched to fp8 training, which got rid of the pattern. But I have to redo all my LoRAs to fix them all. Nothing changed, except the tag is now &lt;lora:LeaningBackAgainst-step0800:1&gt;.<hr />The next pose of the portrait photographers 101 is sitting on the ground and leaning back against ... something. I don't know how they trained Flux, but you can't get that with prompting - the model will always sit in front of the background, not actually touch it. The prompt is simple "leaning back against ..." and I trained it to the strength that is works with a &lt;lora:LeaningBackAgainst-50:1&gt; tag.<hr />Something I started here but had to be placed too in separate LoRAs to work is a finer control over face angles, body angles, camera angles and image area. But starting with this one, the captions are consistent as a preparation, even if they do not (yet) work standalone. The scheme is:Face angles: Full face view | Three-quarter face view | Two-third face view | Profile face viewBody angles: Front angle shot | Three-quarter angle shot | Side angle shot | Three-quarter rear angle shotCamera angle: High camera angle | Eye level shot | Low camera angle | Shoulder level shot | Cowboy shot | Ground level shotImage area: Full-body portrait | Three-quarter-body portrait | Half-body portrait | Quarter-body portraitIt would be great if I get this to work!

Leaning back against

Version 2.0 is trained with fp8, getting rid of the grid pattern sometimes visible in the images.

Files with SHA256: 3988b7bf966c35b3a0413bc5fd91e7484a2c157f61467c4455bc1d07f5265ca9