modelscope
diff --git a/‎diffsynth/models/flux_infiniteyou.py
Copy file name to clipboard
+1Lines changed: 1 addition & 0 deletions b/‎diffsynth/models/flux_infiniteyou.py
Copy file name to clipboard
+1Lines changed: 1 addition & 0 deletions
diff --git a/‎diffsynth/models/flux_value_control.py
Copy file name to clipboard
+2-1Lines changed: 2 additions & 1 deletion b/‎diffsynth/models/flux_value_control.py
Copy file name to clipboard
+2-1Lines changed: 2 additions & 1 deletion
diff --git a/‎diffsynth/models/step1x_connector.py
Copy file name to clipboard
+2-2Lines changed: 2 additions & 2 deletions b/‎diffsynth/models/step1x_connector.py
Copy file name to clipboard
+2-2Lines changed: 2 additions & 2 deletions
diff --git a/‎diffsynth/pipelines/flux_image_new.py
Copy file name to clipboard
+56-24Lines changed: 56 additions & 24 deletions b/‎diffsynth/pipelines/flux_image_new.py
Copy file name to clipboard
+56-24Lines changed: 56 additions & 24 deletions
diff --git a/‎examples/flux/model_inference_low_vram/EliGen.py
Copy file name to clipboard
+148Lines changed: 148 additions & 0 deletions b/‎examples/flux/model_inference_low_vram/EliGen.py
Copy file name to clipboard
+148Lines changed: 148 additions & 0 deletions
diff --git a/‎examples/flux/model_inference_low_vram/FLEX.2-preview.py
Copy file name to clipboard
+51Lines changed: 51 additions & 0 deletions b/‎examples/flux/model_inference_low_vram/FLEX.2-preview.py
Copy file name to clipboard
+51Lines changed: 51 additions & 0 deletions
@@ -104,6 +104,7 @@ def __init__(
     def forward(self, x):
 
         latents = self.latents.repeat(x.size(0), 1, 1)
+        latents = latents.to(dtype=x.dtype, device=x.device)
 
         x = self.proj_in(x)
 
 
@@ -40,7 +40,8 @@ def forward(self, value, dtype):
         emb = self.prefer_proj(value).to(dtype)
         emb = self.prefer_value_embedder(emb).squeeze(0)
         base_embeddings = emb.expand(self.prefer_len, -1)
-        learned_embeddings = base_embeddings + self.positional_embedding
+        positional_embedding = self.positional_embedding.to(dtype=base_embeddings.dtype, device=base_embeddings.device)
+        learned_embeddings = base_embeddings + positional_embedding
         return learned_embeddings
 
     @staticmethod
 
@@ -162,7 +162,7 @@ def timestep_embedding(t, dim, max_period=10000):
     def forward(self, t):
         t_freq = self.timestep_embedding(
             t, self.frequency_embedding_size, self.max_period
-        ).type(self.mlp[0].weight.dtype)  # type: ignore
+        ).type(t.dtype)  # type: ignore
         t_emb = self.mlp(t_freq)
         return t_emb
 
@@ -656,7 +656,7 @@ def forward(self, x,t,mask):
         mask_float = mask.unsqueeze(-1)  # [b, s1, 1]
         x_mean = (x * mask_float).sum(
                 dim=1
-            ) / mask_float.sum(dim=1) * (1 + self.scale_factor)
+            ) / mask_float.sum(dim=1) * (1 + self.scale_factor.to(dtype=x.dtype, device=x.device))
 
         global_out=self.global_proj_out(x_mean)
         encoder_hidden_states = self.S(x,t,mask)
 
@@ -24,7 +24,6 @@
 from .wan_video_new import BasePipeline, ModelConfig, PipelineUnitRunner, PipelineUnit
 from ..lora.flux_lora import FluxLoRALoader, FluxLoraPatcher
 
-from transformers.models.t5.modeling_t5 import T5LayerNorm, T5DenseActDense, T5DenseGatedActDense
 from ..models.flux_dit import RMSNorm
 from ..vram_management import gradient_checkpoint_forward, enable_vram_management, AutoWrappedModule, AutoWrappedLinear
 
@@ -185,22 +184,18 @@ def training_loss(self, **inputs):
         return loss
 
 
-    def enable_vram_management(self, num_persistent_param_in_dit=None, vram_limit=None, vram_buffer=0.5):
-        self.vram_management_enabled = True
-        if num_persistent_param_in_dit is not None:
-            vram_limit = None
-        else:
-            if vram_limit is None:
-                vram_limit = self.get_vram()
-            vram_limit = vram_limit - vram_buffer
-        if self.text_encoder_1 is not None:
-            dtype = next(iter(self.text_encoder_1.parameters())).dtype
+    def _enable_vram_management_with_default_config(self, model, vram_limit):
+        if model is not None:
+            dtype = next(iter(model.parameters())).dtype
             enable_vram_management(
-                self.text_encoder_1,
+                model,
                 module_map = {
                     torch.nn.Linear: AutoWrappedLinear,
                     torch.nn.Embedding: AutoWrappedModule,
                     torch.nn.LayerNorm: AutoWrappedModule,
+                    torch.nn.Conv2d: AutoWrappedModule,
+                    torch.nn.GroupNorm: AutoWrappedModule,
+                    RMSNorm: AutoWrappedModule,
                 },
                 module_config = dict(
                     offload_dtype=dtype,
@@ -212,7 +207,25 @@ def enable_vram_management(self, num_persistent_param_in_dit=None, vram_limit=No
                 ),
                 vram_limit=vram_limit,
             )
+    
+    
+    def enable_vram_management(self, num_persistent_param_in_dit=None, vram_limit=None, vram_buffer=0.5):
+        self.vram_management_enabled = True
+        if num_persistent_param_in_dit is not None:
+            vram_limit = None
+        else:
+            if vram_limit is None:
+                vram_limit = self.get_vram()
+            vram_limit = vram_limit - vram_buffer
+
+        # Default config
+        default_vram_management_models = ["text_encoder_1", "vae_decoder", "vae_encoder", "controlnet", "image_proj_model", "ipadapter", "lora_patcher", "value_controller", "step1x_connector"]
+        for model_name in default_vram_management_models:
+            self._enable_vram_management_with_default_config(getattr(self, model_name), vram_limit)
+
+        # Special config
         if self.text_encoder_2 is not None:
+            from transformers.models.t5.modeling_t5 import T5LayerNorm, T5DenseActDense, T5DenseGatedActDense
             dtype = next(iter(self.text_encoder_2.parameters())).dtype
             enable_vram_management(
                 self.text_encoder_2,
@@ -261,14 +274,18 @@ def enable_vram_management(self, num_persistent_param_in_dit=None, vram_limit=No
                 ),
                 vram_limit=vram_limit,
             )
-        if self.vae_decoder is not None:
-            dtype = next(iter(self.vae_decoder.parameters())).dtype
+        if self.ipadapter_image_encoder is not None:
+            from transformers.models.siglip.modeling_siglip import SiglipVisionEmbeddings, SiglipEncoder, SiglipMultiheadAttentionPoolingHead
+            dtype = next(iter(self.ipadapter_image_encoder.parameters())).dtype
             enable_vram_management(
-                self.vae_decoder,
+                self.ipadapter_image_encoder,
                 module_map = {
+                    SiglipVisionEmbeddings: AutoWrappedModule,
+                    SiglipEncoder: AutoWrappedModule,
+                    SiglipMultiheadAttentionPoolingHead: AutoWrappedModule,
+                    torch.nn.MultiheadAttention: AutoWrappedModule,
                     torch.nn.Linear: AutoWrappedLinear,
-                    torch.nn.Conv2d: AutoWrappedModule,
-                    torch.nn.GroupNorm: AutoWrappedModule,
+                    torch.nn.LayerNorm: AutoWrappedModule,
                 },
                 module_config = dict(
                     offload_dtype=dtype,
@@ -280,14 +297,25 @@ def enable_vram_management(self, num_persistent_param_in_dit=None, vram_limit=No
                 ),
                 vram_limit=vram_limit,
             )
-        if self.vae_encoder is not None:
-            dtype = next(iter(self.vae_encoder.parameters())).dtype
+        if self.qwenvl is not None:
+            from transformers.models.qwen2_5_vl.modeling_qwen2_5_vl import (
+                Qwen2_5_VisionPatchEmbed, Qwen2_5_VLVisionBlock, Qwen2_5_VLPatchMerger,
+                Qwen2_5_VLDecoderLayer, Qwen2_5_VisionRotaryEmbedding, Qwen2_5_VLRotaryEmbedding, Qwen2RMSNorm
+            )
+            dtype = next(iter(self.qwenvl.parameters())).dtype
             enable_vram_management(
-                self.vae_encoder,
+                self.qwenvl,
                 module_map = {
+                    Qwen2_5_VisionPatchEmbed: AutoWrappedModule,
+                    Qwen2_5_VLVisionBlock: AutoWrappedModule,
+                    Qwen2_5_VLPatchMerger: AutoWrappedModule,
+                    Qwen2_5_VLDecoderLayer: AutoWrappedModule,
+                    Qwen2_5_VisionRotaryEmbedding: AutoWrappedModule,
+                    Qwen2_5_VLRotaryEmbedding: AutoWrappedModule,
+                    Qwen2RMSNorm: AutoWrappedModule,
+                    torch.nn.Embedding: AutoWrappedModule,
                     torch.nn.Linear: AutoWrappedLinear,
-                    torch.nn.Conv2d: AutoWrappedModule,
-                    torch.nn.GroupNorm: AutoWrappedModule,
+                    torch.nn.LayerNorm: AutoWrappedModule,
                 },
                 module_config = dict(
                     offload_dtype=dtype,
@@ -774,9 +802,13 @@ def process(self, pipe: FluxImagePipeline, latents, flex_inpaint_image, flex_inp
 
 class FluxImageUnit_InfiniteYou(PipelineUnit):
     def __init__(self):
-        super().__init__(input_params=("infinityou_id_image", "infinityou_guidance"))
+        super().__init__(
+            input_params=("infinityou_id_image", "infinityou_guidance"),
+            onload_model_names=("infinityou_processor",)
+        )
 
     def process(self, pipe: FluxImagePipeline, infinityou_id_image, infinityou_guidance):
+        pipe.load_models_to_device("infinityou_processor")
         if infinityou_id_image is not None:
             return pipe.infinityou_processor.prepare_infinite_you(pipe.image_proj_model, infinityou_id_image, infinityou_guidance, pipe.device)
         else:
@@ -816,7 +848,7 @@ def __init__(self, device="cuda", torch_dtype=torch.bfloat16):
         self.app_320.prepare(ctx_id=0, det_size=(320, 320))
         self.app_160 = FaceAnalysis(name='antelopev2', root=insightface_root_path, providers=['CUDAExecutionProvider', 'CPUExecutionProvider'])
         self.app_160.prepare(ctx_id=0, det_size=(160, 160))
-        self.arcface_model = init_recognition_model('arcface', device=self.device)
+        self.arcface_model = init_recognition_model('arcface', device=self.device).to(torch_dtype)
 
     def _detect_face(self, id_image_cv2):
         face_info = self.app_640.get(id_image_cv2)
 
@@ -0,0 +1,148 @@
+import random
+import torch
+from PIL import Image, ImageDraw, ImageFont
+from diffsynth import download_customized_models
+from diffsynth.pipelines.flux_image_new import FluxImagePipeline, ModelConfig
+from modelscope import dataset_snapshot_download
+
+
+def visualize_masks(image, masks, mask_prompts, output_path, font_size=35, use_random_colors=False):
+    # Create a blank image for overlays
+    overlay = Image.new('RGBA', image.size, (0, 0, 0, 0))
+    
+    colors = [
+        (165, 238, 173, 80),
+        (76, 102, 221, 80),
+        (221, 160, 77, 80),
+        (204, 93, 71, 80),
+        (145, 187, 149, 80),
+        (134, 141, 172, 80),
+        (157, 137, 109, 80),
+        (153, 104, 95, 80),
+        (165, 238, 173, 80),
+        (76, 102, 221, 80),
+        (221, 160, 77, 80),
+        (204, 93, 71, 80),
+        (145, 187, 149, 80),
+        (134, 141, 172, 80),
+        (157, 137, 109, 80),
+        (153, 104, 95, 80),
+    ]
+    # Generate random colors for each mask
+    if use_random_colors:
+        colors = [(random.randint(0, 255), random.randint(0, 255), random.randint(0, 255), 80) for _ in range(len(masks))]
+    
+    # Font settings
+    try:
+        font = ImageFont.truetype("arial", font_size)  # Adjust as needed
+    except IOError:
+        font = ImageFont.load_default(font_size)
+
+    # Overlay each mask onto the overlay image
+    for mask, mask_prompt, color in zip(masks, mask_prompts, colors):
+        # Convert mask to RGBA mode
+        mask_rgba = mask.convert('RGBA')
+        mask_data = mask_rgba.getdata()
+        new_data = [(color if item[:3] == (255, 255, 255) else (0, 0, 0, 0)) for item in mask_data]
+        mask_rgba.putdata(new_data)
+
+        # Draw the mask prompt text on the mask
+        draw = ImageDraw.Draw(mask_rgba)
+        mask_bbox = mask.getbbox()  # Get the bounding box of the mask
+        text_position = (mask_bbox[0] + 10, mask_bbox[1] + 10)  # Adjust text position based on mask position
+        draw.text(text_position, mask_prompt, fill=(255, 255, 255, 255), font=font)
+
+        # Alpha composite the overlay with this mask
+        overlay = Image.alpha_composite(overlay, mask_rgba)
+    
+    # Composite the overlay onto the original image
+    result = Image.alpha_composite(image.convert('RGBA'), overlay)
+    
+    # Save or display the resulting image
+    result.save(output_path)
+
+    return result
+
+def example(pipe, seeds, example_id, global_prompt, entity_prompts):
+    dataset_snapshot_download(dataset_id="DiffSynth-Studio/examples_in_diffsynth", local_dir="./", allow_file_pattern=f"data/examples/eligen/entity_control/example_{example_id}/*.png")
+    masks = [Image.open(f"./data/examples/eligen/entity_control/example_{example_id}/{i}.png").convert('RGB') for i in range(len(entity_prompts))]
+    negative_prompt = "worst quality, low quality, monochrome, zombie, interlocked fingers, Aissist, cleavage, nsfw,"
+    for seed in seeds:
+        # generate image
+        image = pipe(
+            prompt=global_prompt,
+            cfg_scale=3.0,
+            negative_prompt=negative_prompt,
+            num_inference_steps=50,
+            embedded_guidance=3.5,
+            seed=seed,
+            height=1024,
+            width=1024,
+            eligen_entity_prompts=entity_prompts,
+            eligen_entity_masks=masks,
+        )
+        image.save(f"eligen_example_{example_id}_{seed}.png")
+        visualize_masks(image, masks, entity_prompts, f"eligen_example_{example_id}_mask_{seed}.png")
+
+
+pipe = FluxImagePipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="black-forest-labs/FLUX.1-dev", origin_file_pattern="flux1-dev.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
+        ModelConfig(model_id="black-forest-labs/FLUX.1-dev", origin_file_pattern="text_encoder/model.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
+        ModelConfig(model_id="black-forest-labs/FLUX.1-dev", origin_file_pattern="text_encoder_2/", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
+        ModelConfig(model_id="black-forest-labs/FLUX.1-dev", origin_file_pattern="ae.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
+    ],
+)
+pipe.enable_vram_management()
+
+download_from_modelscope = True
+if download_from_modelscope:
+    model_id = "DiffSynth-Studio/Eligen"
+    downloading_priority = ["ModelScope"]
+else:
+    model_id = "modelscope/EliGen"
+    downloading_priority = ["HuggingFace"]
+EliGen_path = download_customized_models(
+    model_id=model_id,
+    origin_file_path="model_bf16.safetensors",
+    local_dir="models/lora/entity_control",
+    downloading_priority=downloading_priority)[0]
+pipe.load_lora(pipe.dit, EliGen_path, alpha=1)
+
+# example 1
+global_prompt = "A breathtaking beauty of Raja Ampat by the late-night moonlight , one beautiful woman from behind wearing a pale blue long dress with soft glow, sitting at the top of a cliff looking towards the beach,pastell light colors, a group of small distant birds flying in far sky, a boat sailing on the sea, best quality, realistic, whimsical, fantastic, splash art, intricate detailed, hyperdetailed, maximalist style, photorealistic, concept art, sharp focus, harmony, serenity, tranquility, soft pastell colors,ambient occlusion, cozy ambient lighting, masterpiece, liiv1, linquivera, metix, mentixis, masterpiece, award winning, view from above\n"
+entity_prompts = ["cliff", "sea", "moon", "sailing boat", "a seated beautiful woman", "pale blue long dress with soft glow"]
+example(pipe, [0], 1, global_prompt, entity_prompts)
+
+# example 2
+global_prompt = "samurai girl wearing a kimono, she's holding a sword  glowing with red flame, her long hair is flowing in the wind, she is looking at a small bird perched on the back of her hand. ultra realist style. maximum image detail. maximum realistic render."
+entity_prompts = ["flowing hair", "sword glowing with red flame", "A cute bird", "blue belt"]
+example(pipe, [0], 2, global_prompt, entity_prompts)
+
+# example 3
+global_prompt = "Image of a neverending staircase up to a mysterious palace in the sky, The ancient palace stood majestically atop a mist-shrouded mountain, sunrise, two traditional monk walk in the stair looking at the sunrise, fog,see-through, best quality, whimsical, fantastic, splash art, intricate detailed, hyperdetailed, photorealistic, concept art, harmony, serenity, tranquility, ambient occlusion, halation, cozy ambient lighting, dynamic lighting,masterpiece, liiv1, linquivera, metix, mentixis, masterpiece, award winning,"
+entity_prompts = ["ancient palace", "stone staircase with railings", "a traditional monk", "a traditional monk"]
+example(pipe, [27], 3, global_prompt, entity_prompts)
+
+# example 4
+global_prompt = "A beautiful girl wearing shirt and shorts in the street,  holding a sign 'Entity Control'"
+entity_prompts = ["A beautiful girl", "sign 'Entity Control'", "shorts", "shirt"]
+example(pipe, [21], 4, global_prompt, entity_prompts)
+
+# example 5
+global_prompt = "A captivating, dramatic scene in a painting that exudes mystery and foreboding. A white sky, swirling blue clouds, and a crescent yellow moon illuminate a solitary woman standing near the water's edge. Her long dress flows in the wind, silhouetted against the eerie glow. The water mirrors the fiery sky and moonlight, amplifying the uneasy atmosphere."
+entity_prompts = ["crescent yellow moon", "a solitary woman", "water", "swirling blue clouds"]
+example(pipe, [0], 5, global_prompt, entity_prompts)
+
+# example 6
+global_prompt = "Snow White and the 6 Dwarfs."
+entity_prompts = ["Dwarf 1", "Dwarf 2", "Dwarf 3", "Snow White", "Dwarf 4", "Dwarf 5", "Dwarf 6"]
+example(pipe, [8], 6, global_prompt, entity_prompts)
+
+# example 7, same prompt with different seeds
+seeds = range(5, 9)
+global_prompt = "A beautiful woman wearing white dress, holding a mirror, with a warm light background;"
+entity_prompts = ["A beautiful woman", "mirror", "necklace", "glasses", "earring", "white dress", "jewelry headpiece"]
+example(pipe, seeds, 7, global_prompt, entity_prompts)
@@ -0,0 +1,51 @@
+import torch
+from diffsynth.pipelines.flux_image_new import FluxImagePipeline, ModelConfig
+from diffsynth.controlnets.processors import Annotator
+import numpy as np
+from PIL import Image
+
+
+pipe = FluxImagePipeline.from_pretrained(
+    torch_dtype=torch.bfloat16,
+    device="cuda",
+    model_configs=[
+        ModelConfig(model_id="ostris/Flex.2-preview", origin_file_pattern="Flex.2-preview.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
+        ModelConfig(model_id="black-forest-labs/FLUX.1-dev", origin_file_pattern="text_encoder/model.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
+        ModelConfig(model_id="black-forest-labs/FLUX.1-dev", origin_file_pattern="text_encoder_2/", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
+        ModelConfig(model_id="black-forest-labs/FLUX.1-dev", origin_file_pattern="ae.safetensors", offload_device="cpu", offload_dtype=torch.float8_e4m3fn),
+    ],
+)
+pipe.enable_vram_management()
+
+image = pipe(
+    prompt="portrait of a beautiful Asian girl, long hair, red t-shirt, sunshine, beach",
+    num_inference_steps=50, embedded_guidance=3.5,
+    seed=0
+)
+image.save(f"image_1.jpg")
+
+mask = np.zeros((1024, 1024, 3), dtype=np.uint8)
+mask[200:400, 400:700] = 255
+mask = Image.fromarray(mask)
+mask.save(f"image_mask.jpg")
+
+inpaint_image = image
+
+image = pipe(
+    prompt="portrait of a beautiful Asian girl with sunglasses, long hair, red t-shirt, sunshine, beach",
+    num_inference_steps=50, embedded_guidance=3.5,
+    flex_inpaint_image=inpaint_image, flex_inpaint_mask=mask,
+    seed=4
+)
+image.save(f"image_2_new.jpg")
+
+control_image = Annotator("canny")(image)
+control_image.save("image_control.jpg")
+
+image = pipe(
+    prompt="portrait of a beautiful Asian girl with sunglasses, long hair, yellow t-shirt, sunshine, beach",
+    num_inference_steps=50, embedded_guidance=3.5,
+    flex_control_image=control_image,
+    seed=4
+)
+image.save(f"image_3_new.jpg")