refine examples

diffsynth/pipelines/wan_video_new.py

@@ -169,7 +169,7 @@ class ModelConfig:
     def download_if_necessary(self, local_model_path="./models", skip_download=False, use_usp=False):
         if self.path is None:
             # Check model_id and origin_file_pattern
-            if self.model_id is None or self.origin_file_pattern is None:
+            if self.model_id is None:
                 raise ValueError(f"""No valid model files. Please use `ModelConfig(path="xxx")` or `ModelConfig(model_id="xxx/yyy", origin_file_pattern="zzz")`.""")
             
             # Skip if not in rank 0
@@ -178,7 +178,11 @@ class ModelConfig:
                 skip_download = dist.get_rank() != 0
                 
             # Check whether the origin path is a folder
-            if isinstance(self.origin_file_pattern, str) and self.origin_file_pattern.endswith("/"):
+            if self.origin_file_pattern is None:
+                self.origin_file_pattern = ""
+                allow_file_pattern = None
+                is_folder = True
+            elif isinstance(self.origin_file_pattern, str) and self.origin_file_pattern.endswith("/"):
                 allow_file_pattern = self.origin_file_pattern + "*"
                 is_folder = True
             else:

examples/flux/FLUX.1-dev-IP-Adapter.py

@@ -1,10 +1,7 @@
 import torch
-from PIL import Image
-from diffsynth import save_video, VideoData, download_models
 from diffsynth.pipelines.flux_image_new import FluxImagePipeline, ModelConfig
-from modelscope import dataset_snapshot_download
 
-#TODO: repalce the local path with model_id
+
 pipe = FluxImagePipeline.from_pretrained(
     torch_dtype=torch.bfloat16,
     device="cuda",
@@ -14,16 +11,14 @@ pipe = FluxImagePipeline.from_pretrained(
         ModelConfig(model_id="black-forest-labs/FLUX.1-dev", origin_file_pattern="text_encoder_2/"),
         ModelConfig(model_id="black-forest-labs/FLUX.1-dev", origin_file_pattern="ae.safetensors"),
         ModelConfig(model_id="InstantX/FLUX.1-dev-IP-Adapter", origin_file_pattern="ip-adapter.bin"),
-        ModelConfig(path="models/IpAdapter/InstantX/FLUX.1-dev-IP-Adapter/image_encoder")
+        ModelConfig(model_id="google/siglip-so400m-patch14-384"),
     ],
 )
 
-seed = 42
 origin_prompt = "a rabbit in a garden, colorful flowers"
-image = pipe(prompt=origin_prompt, height=1280, width=960, seed=seed)
+image = pipe(prompt=origin_prompt, height=1280, width=960, seed=42)
 image.save("style image.jpg")
 
-torch.manual_seed(seed)
-image = pipe(prompt="A piggy", height=1280, width=960, seed=seed,
+image = pipe(prompt="A piggy", height=1280, width=960, seed=42,
     ipadapter_images=[image], ipadapter_scale=0.7)
 image.save("A piggy.jpg")

examples/flux/Step1X-Edit.py

@@ -1,13 +1,9 @@
 import torch
 from diffsynth.pipelines.flux_image_new import FluxImagePipeline, ModelConfig
-from modelscope import snapshot_download
 from PIL import Image
 import numpy as np
 
 
-snapshot_download("Qwen/Qwen2.5-VL-7B-Instruct", cache_dir="./models")
-snapshot_download("stepfun-ai/Step1X-Edit", cache_dir="./models")
-
 pipe = FluxImagePipeline.from_pretrained(
     torch_dtype=torch.bfloat16,
     device="cuda",
@@ -18,27 +14,19 @@ pipe = FluxImagePipeline.from_pretrained(
     ],
 )
 
-
-pipe.enable_vram_management()
-
 image = Image.fromarray(np.zeros((1248, 832, 3), dtype=np.uint8) + 255)
 image = pipe(
     prompt="draw red flowers in Chinese ink painting style",
     step1x_reference_image=image,
     width=832, height=1248, cfg_scale=6,
-    seed=1,
-    rand_device='cuda'
+    seed=1, rand_device='cuda'
 )
 image.save("image_1.jpg")
 
-
-
 image = pipe(
     prompt="add more flowers in Chinese ink painting style",
     step1x_reference_image=image,
     width=832, height=1248, cfg_scale=6,
-    seed=2,
-    rand_device='cuda'
+    seed=2, rand_device='cuda'
 )
 image.save("image_2.jpg")
-