update separate_cfg_infer,

app.py

@@ -11,7 +11,7 @@ pipe = OmniGenPipeline.from_pretrained(
 
 @spaces.GPU(duration=180)
 # 示例处理函数：生成图像
-def generate_image(text, img1, img2, img3, height, width, guidance_scale, img_guidance_scale, inference_steps, seed):
+def generate_image(text, img1, img2, img3, height, width, guidance_scale, img_guidance_scale, inference_steps, seed, separate_cfg_infer):
     input_images = [img1, img2, img3]
     # 去除 None
     input_images = [img for img in input_images if img is not None]
@@ -29,6 +29,7 @@ def generate_image(text, img1, img2, img3, height, width, guidance_scale, img_gu
         separate_cfg_infer=True, # set False can speed up the inference process
         use_kv_cache=False,
         seed=seed,
+        separate_cfg_infer=separate_cfg_infer,
     )
     img = output[0]
     return img
@@ -57,6 +58,7 @@ def get_example():
             1.6,
             50,
             0,
+            False,
         ],
         [
             "The woman in <img><|image_1|></img> waves her hand happily in the crowd",
@@ -69,6 +71,7 @@ def get_example():
             1.9,
             50,
             128,
+            False,
         ],
         [
             "A man in a black shirt is reading a book. The man is the right man in <img><|image_1|></img>.",
@@ -81,6 +84,7 @@ def get_example():
             1.6,
             50,
             0,
+            False,
         ],
         [
             "Two woman are raising fried chicken legs in a bar. A woman is <img><|image_1|></img>. The other woman is <img><|image_2|></img>.",
@@ -93,6 +97,7 @@ def get_example():
             1.8,
             50,
             168,
+            False,
         ],
         [
             "A man and a short-haired woman with a wrinkled face are standing in front of a bookshelf in a library. The man is the man in the middle of <img><|image_1|></img>, and the woman is oldest woman in <img><|image_2|></img>",
@@ -105,6 +110,7 @@ def get_example():
             1.6,
             50,
             60,
+            False,
         ],
         [
             "A man and a woman are sitting at a classroom desk. The man is the man with yellow hair in <img><|image_1|></img>. The woman is the woman on the left of <img><|image_2|></img>",
@@ -117,6 +123,7 @@ def get_example():
             1.8,
             50,
             66,
+            False,
         ],
         [
             "The flower <img><|image_1|><\/img> is placed in the vase which is in the middle of <img><|image_2|><\/img> on a wooden table of a living room",
@@ -129,6 +136,7 @@ def get_example():
             1.6,
             50,
             0,
+            False,
         ],
         [
             "<img><|image_1|><img>\n Remove the woman's earrings. Replace the mug with a clear glass filled with sparkling iced cola.",
@@ -141,6 +149,7 @@ def get_example():
             1.6,
             50,
             222,
+            False,
         ],
         [
             "Detect the skeleton of human in this image: <img><|image_1|></img>.",
@@ -153,6 +162,7 @@ def get_example():
             1.6,
             50,
             0,
+            False,
         ],
         [
             "Generate a new photo using the following picture and text as conditions: <img><|image_1|><img>\n A young boy is sitting on a sofa in the library, holding a book. His hair is neatly combed, and a faint smile plays on his lips, with a few freckles scattered across his cheeks. The library is quiet, with rows of shelves filled with books stretching out behind him.",
@@ -165,6 +175,7 @@ def get_example():
             1.6,
             50,
             42,
+            False,
         ],
         [
             "Following the pose of this image <img><|image_1|><img>, generate a new photo: A young boy is sitting on a sofa in the library, holding a book. His hair is neatly combed, and a faint smile plays on his lips, with a few freckles scattered across his cheeks. The library is quiet, with rows of shelves filled with books stretching out behind him.",
@@ -177,6 +188,7 @@ def get_example():
             1.6,
             50,
             123,
+            False,
         ],
         [
             "Following the depth mapping of this image <img><|image_1|><img>, generate a new photo: A young girl is sitting on a sofa in the library, holding a book. His hair is neatly combed, and a faint smile plays on his lips, with a few freckles scattered across his cheeks. The library is quiet, with rows of shelves filled with books stretching out behind him.",
@@ -189,6 +201,7 @@ def get_example():
             1.6,
             50,
             1,
+            False,
         ],
         [
             "<img><|image_1|><\/img> What item can be used to see the current time? Please remove it.",
@@ -201,6 +214,7 @@ def get_example():
             1.6,
             50,
             0,
+            False,
         ],
         [
             "According to the following examples, generate an output for the input.\nInput: <img><|image_1|></img>\nOutput: <img><|image_2|></img>\n\nInput: <img><|image_3|></img>\nOutput: ",
@@ -213,12 +227,13 @@ def get_example():
             1.6,
             50,
             1,
+            False,
         ],
     ]
     return case
 
-def run_for_examples(text, img1, img2, img3, height, width, guidance_scale, img_guidance_scale, inference_steps, seed):    
-    return generate_image(text, img1, img2, img3, height, width, guidance_scale, img_guidance_scale, inference_steps, seed)
+def run_for_examples(text, img1, img2, img3, height, width, guidance_scale, img_guidance_scale, inference_steps, seed, separate_cfg_infer,):    
+    return generate_image(text, img1, img2, img3, height, width, guidance_scale, img_guidance_scale, inference_steps, seed, separate_cfg_infer,)
 
 description = """
 OmniGen is a unified image generation model that you can use to perform various tasks, including but not limited to text-to-image generation, subject-driven generation, Identity-Preserving Generation, and image-conditioned generation.
@@ -235,6 +250,8 @@ Tips:
 - For image editing tasks, we recommend placing the image before the editing instruction. For example, use `<img><|image_1|></img> remove suit`, rather than `remove suit <img><|image_1|></img>`.
 """
 
+separate_cfg_infer_arg = False
+
 # Gradio 接口
 with gr.Blocks() as demo:
     gr.Markdown("# OmniGen: Unified Image Generation [paper](https://arxiv.org/abs/2409.11340) [code](https://github.com/VectorSpaceLab/OmniGen)")
@@ -277,6 +294,10 @@ with gr.Blocks() as demo:
                 label="Seed", minimum=0, maximum=2147483647, value=42, step=1
             )
 
+            separate_cfg_infer = gr.Checkbox(
+                label="separate_cfg_infer", info="enable separate cfg infer"
+            )
+
             # 生成按钮
             generate_button = gr.Button("Generate Image")
 
@@ -298,6 +319,7 @@ with gr.Blocks() as demo:
             img_guidance_scale_input,
             num_inference_steps,
             seed_input,
+            separate_cfg_infer,
         ],
         outputs=output_image,
     )
@@ -316,6 +338,7 @@ with gr.Blocks() as demo:
             img_guidance_scale_input,
             num_inference_steps,
             seed_input,
+            separate_cfg_infer,
         ],
         outputs=output_image,
     )