Update app.py

app.py

@@ -297,25 +297,24 @@ def generate_images(prompt_mash, steps, seed, cfg_scale, width, height, progress
     print("Generating multiple images...")
     pipe.to("cuda")
     images = []
-
     for _ in range(4):  # Generate 4 images
         seed = random.randint(0, MAX_SEED)
         generator = torch.Generator(device="cuda").manual_seed(seed)
         with calculateDuration("Generating image"):
-            for img in pipe.flux_pipe_call_that_returns_an_iterable_of_images(
-                prompt=prompt_mash,
-                num_inference_steps=steps,
-                guidance_scale=cfg_scale,
-                width=width,
-                height=height,
-                generator=generator,
-                joint_attention_kwargs={"scale": 1.0},
-                output_type="pil",
-                good_vae=good_vae,
-            ):
-                images.append((img, seed))  # Store image and its seed
-                break  # Only take the first generated image
-
+            img = next(
+                pipe.flux_pipe_call_that_returns_an_iterable_of_images(
+                    prompt=prompt_mash,
+                    num_inference_steps=steps,
+                    guidance_scale=cfg_scale,
+                    width=width,
+                    height=height,
+                    generator=generator,
+                    joint_attention_kwargs={"scale": 1.0},
+                    output_type="pil",
+                    good_vae=good_vae,
+                )
+            )
+            images.append((img, seed))
     return images
 
 #def generate_image_to_image(prompt_mash, image_input_path, image_strength, steps, cfg_scale, width, height, seed):
@@ -334,7 +333,7 @@ def generate_images(prompt_mash, steps, seed, cfg_scale, width, height, progress
 #        joint_attention_kwargs={"scale": 1.0},
 #        output_type="pil",
 #    ).images[0]
-    return img
+#    return img
 
 @spaces.GPU(duration=75)
 def run_lora_multi(prompt, cfg_scale, steps, selected_indices, lora_scale_1, lora_scale_2, randomize_seed, seed, width, height, loras_state, progress=gr.Progress(track_tqdm=True)):