Upload 7 files

.gitattributes

@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+assets/images/5.jpeg filter=lfs diff=lfs merge=lfs -text
+assets/images/802f12c980c54aa18e6c3d75d3d6bafe1734588260122.png filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,15 @@
+---
+title: Wan2.1
+emoji: 💻
+colorFrom: indigo
+colorTo: indigo
+sdk: gradio
+sdk_version: 5.17.1
+app_file: app.py
+pinned: false
+license: mit
+short_description: 'Wan: Open and Advanced Large-Scale Video Generative Models'
+---
+
+Online Demo: https://imagetovideoaifree.com/wanx21
+

app.py

@@ -0,0 +1,307 @@
+import copy
+import os
+import random
+
+os.system('pip install dashscope')
+import gradio as gr
+import dashscope
+from dashscope import VideoSynthesis
+from examples import t2v_examples, i2v_examples
+import time
+
+DASHSCOPE_API_KEY = os.getenv('DASHSCOPE_API_KEY')
+dashscope.api_key = DASHSCOPE_API_KEY
+
+KEEP_SUCCESS_TASK = 3600 * 10
+KEEP_RUNING_TASK = 3600 * 2
+# the total running task number in 1800 seconds
+LIMIT_RUNING_TASK = 10
+
+def t2v_generation(prompt, resolution, watermark_wan, seed = -1):
+    seed = seed if seed >= 0 else random.randint(0, 2147483647)
+    if not allow_task_num():
+        gr.Info(f"Warning: The number of running tasks is too large, please wait for a while.")
+        return None, gr.Button(visible=True)
+    try:
+        rsp = VideoSynthesis.call(model="wanx2.1-t2v-plus", prompt=prompt, seed=seed,
+                                  watermark_wanx=watermark_wan, size=resolution)
+        video_url = rsp.output.video_url
+        return video_url, gr.Button(visible=True)
+    except Exception as e:
+        gr.Warning(f"Warning: {e}")
+        return None, gr.Button(visible=True)
+
+def t2v_generation_async(prompt, size, watermark_wan, seed = -1):
+    print(seed)
+    seed = seed if seed >= 0 else random.randint(0, 2147483647)
+    print(seed)
+    if not allow_task_num():
+        gr.Info(f"Warning: The number of running tasks is too large, please wait for a while.")
+        return None, False, gr.Button(visible=True)
+    try:
+        rsp = VideoSynthesis.async_call(model="wanx2.1-t2v-plus",
+                                        prompt=prompt,
+                                        size=size,
+                                        seed=seed,
+                                        watermark_wanx=watermark_wan)
+        task_id = rsp.output.task_id
+        status = False
+        return task_id, status, gr.Button(visible=False)
+    except Exception as e:
+        gr.Warning(f"Warning: {e}")
+        return None, True, gr.Button()
+
+def i2v_generation(prompt, image, watermark_wan, seed = -1):
+    seed = seed if seed >= 0 else random.randint(0, 2147483647)
+    video_url = None
+    try:
+        rsp = VideoSynthesis.call(model="wanx2.1-i2v-plus", prompt=prompt, img_url= image,
+                                  seed = seed,
+                                  watermark_wanx=watermark_wan
+                                  )
+        video_url = rsp.output.video_url
+    except Exception as e:
+        gr.Warning(f"Warning: {e}")
+    return video_url
+
+def i2v_generation_async(prompt, image, watermark_wan, seed = -1):
+    seed = seed if seed >= 0 else random.randint(0, 2147483647)
+    if not allow_task_num():
+        gr.Info(f"Warning: The number of running tasks is too large, please wait for a while.")
+        return "", None, gr.Button(visible=True)
+    try:
+        # 检查image是否为URL字符串
+        if isinstance(image, str) and (image.startswith('http://') or image.startswith('https://')):
+            img_url = image
+        else:
+            img_url = image  # 上传的文件路径
+            
+        rsp = VideoSynthesis.async_call(model="wanx2.1-i2v-plus", prompt=prompt, seed=seed,
+                                        img_url=img_url, watermark_wanx=watermark_wan)
+        print(rsp)
+        task_id = rsp.output.task_id
+        status = False
+        return task_id, status, gr.Button(visible=False)
+    except Exception as e:
+        gr.Warning(f"Warning: {e}")
+        return "", None, gr.Button()
+
+def get_result_with_task_id(task_id):
+    if task_id == "": return True, None
+    try:
+        rsp = VideoSynthesis.fetch(task = task_id)
+        print(rsp)
+        if rsp.output.task_status == "FAILED":
+            gr.Info(f"Warning: task running {rsp.output.task_status}")
+            status = True
+            video_url = None
+        else:
+            video_url = rsp.output.video_url
+            video_url = video_url if video_url != "" else None
+            status = video_url is not None
+    except:
+        video_url = None
+        status = False
+    return status, None if video_url=="" else video_url
+    # return True, "https://dashscope-result-wlcb.oss-cn-wulanchabu.aliyuncs.com/1d/f8/20250220/e7d3f375/ccc590a2-7e90-4d92-84bc-22668db42979.mp4?Expires=1740137152&OSSAccessKeyId=LTAI5tQZd8AEcZX6KZV4G8qL&Signature=i3S3jA5FY6XYfvzZNHnvQiPzZSw%3D"
+task_status = {}
+
+def allow_task_num():
+    num = 0
+    for task_id in task_status:
+        if not task_status[task_id]["status"] and task_status[task_id]["time"] + 1800 > time.time():
+            num += 1
+    return num < LIMIT_RUNING_TASK
+def clean_task_status():
+    # clean the task over 1800 seconds
+    for task_id in copy.deepcopy(task_status):
+        if task_id == "": continue
+        # finished task, keep 3600 seconds
+        if task_status[task_id]["status"]:
+            if task_status[task_id]["time"] + KEEP_SUCCESS_TASK < time.time():
+                task_status.pop(task_id)
+        else:
+            # clean the task over 3600 * 2 seconds
+            if task_status[task_id]["time"] + KEEP_RUNING_TASK < time.time():
+                task_status.pop(task_id)
+
+def cost_time(task_id):
+    if task_id in task_status and not task_status[task_id]["status"]:
+        et = time.time() - task_status[task_id]["time"]
+        return f"{et:.2f}"
+    else:
+        return gr.Textbox()
+def get_process_bar(task_id, status):
+    clean_task_status()
+    if task_id not in task_status:
+        task_status[task_id] = {
+            "value": 0 if not task_id == "" else 100,
+            "status": status if not task_id == "" else True,
+            "time": time.time(),
+            "url": None
+        }
+    if not task_status[task_id]["status"]:
+        # only when > 50% do check status
+        if task_status[task_id]["value"] >= 10 and task_status[task_id]["value"] % 5 == 0:
+            status, video_url = get_result_with_task_id(task_id)
+        else:
+            status, video_url = False, None
+        task_status[task_id]["status"] = status
+        task_status[task_id]["url"] = video_url
+    if task_status[task_id]["status"]:
+        task_status[task_id]["value"] = 100
+    else:
+        task_status[task_id]["value"] += 1
+    if task_status[task_id]["value"] >= 100 and not task_status[task_id]["status"]:
+        task_status[task_id]["value"] = 95
+    # print(task_id, task_status[task_id], task_status)
+    value = task_status[task_id]["value"]
+    return gr.Slider(label= f"({value}%)Generating" if value%2==1 else f"({value}%)Generating.....", value=value)
+
+
+with gr.Blocks() as demo:
+    
+    gr.HTML("""
+               <div style="text-align: center; font-size: 32px; font-weight: bold; margin-bottom: 20px;">
+                   Wan2.1: Open and Advanced Large-Scale Video Generative Models
+               </div>
+               <div style="text-align: center;">
+                   <a href="https://github.com/Wan-Video/Wan2.1">Code</a> |
+                   <a href="https://huggingface.co/Wan-AI">Huggingface</a> |
+                   <a href="https://modelscope.cn/organization/Wan-AI">Modelscope</a>
+               </div>
+               
+               """)
+    task_id = gr.State(value="")
+    status = gr.State(value=False)
+    task = gr.State(value="t2v")
+    with gr.Row():
+        with gr.Column():
+            with gr.Row():
+                with gr.Tabs():
+                    # Text to Video Tab
+                    with gr.TabItem("Text to Video") as t2v_tab:
+                        with gr.Row():
+                            txt2vid_prompt = gr.Textbox(
+                                label="Prompt",
+                                placeholder="Describe the video you want to generate",
+                                lines=19,
+                            )
+                        with gr.Row():
+                            resolution = gr.Dropdown(
+                                label="Resolution",
+                                choices=["1280*720", "960*960", "720*1280", "1088*832", "832*1088"],
+                                value="1280*720",
+                            )
+                        with gr.Row():
+                            run_t2v_button = gr.Button("Generate Video")
+                    # Image to Video Tab
+                    with gr.TabItem("Image to Video") as i2v_tab:
+                        with gr.Row():
+                            with gr.Column():
+                                img_input_type = gr.Radio(
+                                    ["上传图片", "图片URL"], 
+                                    label="输入方式", 
+                                    value="上传图片"
+                                )
+                                
+                                with gr.Group(visible=True) as upload_image_group:
+                                    img2vid_image = gr.Image(
+                                        type="filepath",
+                                        label="上传输入图片",
+                                        elem_id="image_upload",
+                                    )
+                                
+                                with gr.Group(visible=False) as url_image_group:
+                                    img2vid_url = gr.Textbox(
+                                        label="图片URL",
+                                        placeholder="输入图片的URL地址",
+                                        value="",
+                                    )
+                                
+                                img2vid_prompt = gr.Textbox(
+                                    label="Prompt",
+                                    placeholder="Describe the video you want to generate",
+                                    value="",
+                                    lines=5,
+                                )
+                        with gr.Row():
+                            run_i2v_button = gr.Button("生成视频")
+        with gr.Column():
+            with gr.Row():
+                result_gallery = gr.Video(label='Generated Video',
+                                          interactive=False,
+                                          height=500)
+            with gr.Row():
+                    watermark_wan = gr.Checkbox(label="Watermark", value=True, visible=True, container=False)
+                    seed = gr.Number(label="Seed", value=-1, container=True)
+                    cost_time = gr.Number(label="Cost Time(secs)", value=cost_time, interactive=False,
+                                           every=10, inputs=[task_id], container=True)
+                    process_bar = gr.Slider(show_label=True, label="", value=get_process_bar, maximum=100,
+                      interactive=True, every=10, inputs=[task_id, status], container=True)
+
+    fake_video = gr.Video(label='Examples', visible=False, interactive=False)
+    with gr.Row(visible=True) as t2v_eg:
+        gr.Examples(t2v_examples,
+                    inputs=[txt2vid_prompt, result_gallery],
+                    outputs=[result_gallery])
+
+    with gr.Row(visible=False) as i2v_eg:
+        gr.Examples(i2v_examples,
+                    inputs=[img2vid_prompt, img2vid_image, result_gallery],
+                    outputs=[result_gallery])
+
+    def process_change(task_id, task):
+        status = task_status[task_id]["status"]
+        if status:
+            video_url = task_status[task_id]["url"]
+            ret_t2v_btn = gr.Button(visible=True) if task == 't2v' else gr.Button()
+            ret_i2v_btn = gr.Button(visible=True) if task == 'i2v' else gr.Button()
+            return gr.Video(value=video_url), ret_t2v_btn, ret_i2v_btn
+        return gr.Video(value=None), gr.Button(), gr.Button()
+
+    process_bar.change(process_change, inputs=[task_id, task],
+                       outputs=[result_gallery, run_t2v_button, run_i2v_button])
+
+
+    def switch_i2v_tab():
+        return gr.Row(visible=False), gr.Row(visible=True), "i2v"
+
+    def switch_t2v_tab():
+        return gr.Row(visible=True), gr.Row(visible=False), "t2v"
+
+    i2v_tab.select(switch_i2v_tab, outputs=[t2v_eg, i2v_eg, task])
+    t2v_tab.select(switch_t2v_tab, outputs=[t2v_eg, i2v_eg, task])
+
+    run_t2v_button.click(
+        fn=t2v_generation_async,
+        inputs=[txt2vid_prompt, resolution, watermark_wan, seed],
+        outputs=[task_id, status, run_t2v_button],
+    )
+
+    def switch_input_type(input_type):
+        if input_type == "上传图片":
+            return gr.Group(visible=True), gr.Group(visible=False)
+        else:
+            return gr.Group(visible=False), gr.Group(visible=True)
+            
+    img_input_type.change(
+        fn=switch_input_type,
+        inputs=[img_input_type],
+        outputs=[upload_image_group, url_image_group],
+    )
+
+    def i2v_generation_async_wrapper(prompt, image, image_url, input_type, watermark_wan, seed=-1):
+        if input_type == "上传图片":
+            return i2v_generation_async(prompt, image, watermark_wan, seed)
+        else:
+            return i2v_generation_async(prompt, image_url, watermark_wan, seed)
+
+    run_i2v_button.click(
+        fn=i2v_generation_async_wrapper,
+        inputs=[img2vid_prompt, img2vid_image, img2vid_url, img_input_type, watermark_wan, seed],
+        outputs=[task_id, status, run_i2v_button],
+    )
+
+demo.queue(max_size=10)
+demo.launch()

examples.py

@@ -0,0 +1,23 @@
+t2v_examples = [
+    [
+        "在台风肆虐的街头，一只身着亮黄雨衣、背负巨大天使翅膀的橙色小猫，驾驶踏板摩托车勇敢穿行。8K画质下，小猫眼神灵动，毛发细腻，雨衣与头盔色彩鲜明，与灰暗背景形成强烈对比。城市灯光映照在路面积水上，增添一抹柔和，小猫的微笑和闪烁的大眼睛仿佛能驱散所有阴霾，营造出温馨奇幻的氛围，让人仿佛进入了一个充满魔力的梦境。",
+        "./assets/videos/2.webm"
+    ],
+    [
+        "A little dog wearing headphones is riding a colorful scooter across a lush green field, with a basket filled with various fresh vegetables, resembling a moving pastoral landscape painting. Uniquely, a relaxed little cat is perched on the dog's back, and the two of them work together harmoniously, enjoying the thrill of the speedy ride. In the background, rolling mountains and a blue sky decorated with fluffy white clouds create a soothing and delightful scene. The dog's focused expression and the cat's comfortable posture together illustrate this extraordinary joy of riding.",
+        "./assets/videos/1.webm"
+    ],
+]
+
+i2v_examples = [
+    [
+        "一只水母缓缓漂浮在海洋中。水母的伞形体态展现出柔和的蓝色和紫色，中央散发出温暖的金黄色光芒，仿佛在闪烁着神秘的光辉。长长的触手如丝带般轻盈地摇曳",
+        "./assets/images/5.jpeg",
+        "./assets/videos/3.webm"
+    ],
+    [
+        "The cat is moving around, and the flag in the background is waving along with the clouds.",
+        "./assets/images/802f12c980c54aa18e6c3d75d3d6bafe1734588260122.png",
+        "./assets/videos/4.webm"
+    ]
+]

gitattributes

@@ -0,0 +1,35 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

requirements.txt

@@ -0,0 +1,4 @@
+dashscope>=0.10.0
+gradio>=3.50.2
+pillow>=9.0.0
+numpy>=1.22.0