Update app.py

app.py

@@ -27,6 +27,8 @@ from insightface.app import FaceAnalysis
 from pipeline_stable_diffusion_xl_instantid_img2img import StableDiffusionXLInstantIDImg2ImgPipeline, draw_kps
 from controlnet_aux import ZoeDetector
 
+from compel import Compel, ReturnedEmbeddingsType
+
 with open("sdxl_loras.json", "r") as file:
     data = json.load(file)
     sdxl_loras_raw = [
@@ -107,6 +109,9 @@ pipe = StableDiffusionXLInstantIDImg2ImgPipeline.from_pretrained("rubbrband/albe
                                                                  vae=vae,
                                                                  controlnet=[identitynet, zoedepthnet],
                                                                  torch_dtype=torch.float16)
+
+compel = Compel(tokenizer=[pipe.tokenizer, pipeline.tokenizer_2] , text_encoder=[pipe.text_encoder, pipe.text_encoder_2], returned_embeddings_type=ReturnedEmbeddingsType.PENULTIMATE_HIDDEN_STATES_NON_NORMALIZED, requires_pooled=[False, True])
+
 pipe.scheduler = DPMSolverMultistepScheduler.from_config(pipe.scheduler.config, use_karras_sigmas=True)
 pipe.load_ip_adapter_instantid(face_adapter)
 pipe.set_ip_adapter_scale(0.8)
@@ -268,10 +273,14 @@ def run_lora(face_image, prompt, negative, lora_scale, selected_state, face_stre
                 pipe.unload_textual_inversion()
                 pipe.load_textual_inversion(state_dict_embedding["text_encoders_0"], token=["<s0>", "<s1>"], text_encoder=pipe.text_encoder, tokenizer=pipe.tokenizer)
                 pipe.load_textual_inversion(state_dict_embedding["text_encoders_1"], token=["<s0>", "<s1>"], text_encoder=pipe.text_encoder_2, tokenizer=pipe.tokenizer_2)
-                
+
+    conditioning, pooled = compel(prompt)
+    negative_conditioning, negative_pooled = compel(negative)
     image = pipe(
-        prompt=prompt,
-        negative_prompt=negative,
+        prompt_embeds=conditioning,
+        pooled_prompt_embeds=pooled,
+        negative_prompt_embeds=negative_conditioning,
+        negative_pooled_prompt_embeds=negative_pooled,
         width=1024,
         height=1024,
         image_embeds=face_emb,