Prompt Words

  <lora:xi gundam lora:0.7>  <lora:iron man:0.3> gundam iron man robotic realism in a future city background  <lora:xi gundam lora:0.2>  <lora:pensketch_lora_v2.3:0.2>  <lora:iron man:0.6> gundam iron man black and white pen sketch   <lora:xi gundam lora:0.4>  <lora:iron man:0.3> gundam iron man robotic realism in a future city background <lora:blindbox:0.3> blindbox   <lora:xi gundam lora:0.4>  <lora:iron man:0.3> gundam iron man robotic realism in a future city background <lora:blindbox:0.3> blindbox black venom black venom black venom black venom black venom black venom    <lora:xi gundam lora:0.4>  <lora:iron man:0.3>  <lora:blindbox:0.3> blindbox robotic iron man dragon dragon dragon dragon dragon dragon dragon dragon dragon dragon dragon dragon dragon dragon dragon dragon dragon dragon dragon dragon 

  <lora:SingleLineArt-LORA:0.5> <lora:flat design playfulWhimsyLineless_v2:0.5>  SingleLineArt flat design man and dog b&w

  abstract abstract abstract abstract 2d illusion galaxy<lora:vangogh:0.25>  <lora:2D illustration:0.25> van gogh style sea <lora:purehay:0.25> <lora:picasso:0.25> purehay reading books

 one line art one line art one line art one line art one line art lady abstract <lora:2D illustration:0.25> <lora:picasso:0.25> <lora:purehay:0.25> <lora:handdraw_minimalist_lineart:0.25>

 abstract abstract abstract abstract 2d illusion nude lady<lora:vangogh:0.25>  <lora:2D illustration:0.25> van gogh style lady leg <lora:purehay:0.25> <lora:picasso:0.25> purehay reading books japanese abstract abstract abstract abstract 2d illusion nude lady<lora:vangogh:0.25>  van gogh style lady leg <lora:purehay:0.25> <lora:picasso:0.25> purehay reading books   <lora:irasutoya:0.25> irasutoya

  "controlnet_input_image": [img], "controlnet_module": 'openpose', "controlnet_model": 'control_sd15_openpose [fef5e48e]', This is how json objects are formated and those that are necessary. My setup works fine this way with depth, pose and scribble. Img is a base64 string for the image that goes into the ControlNet UI on the SD WebUI. EDIT: control img can be added but procrastinating on it. If interested in doing so, just need to encode the img in the second index of the output that is in numpy array format. Good. Small advice: Move API stuff to another file, like  api.py Maybe we should add some examples in the readme.