VACE 14B-Vensheng Video (Automatically Expand Prompt Words)
5.0
0 reviewsDescription
🚀Edit& Run it online (RunningHub):
https://www.runninghub.cn/post/1922918246369083394/aiDetail/?utm_source=openar
RunningHub – Highly Reliable Cloud-Based ComfyUI, Edit and Run Workflows Online, no local installation required. Powered by RTX 4090 GPUs for faster performance and fully error-free node operation.
----------------------------------------------------------
🚀 在线编辑运行(RunningHub):
https://www.runninghub.cn/ai-detail/1922975290036056065/?utm_source=openar
【RunningHub - 超高可用性云端ComfyUI,在线编辑运行工作流,标配RTX4090带来更快运行速度,节点全不报红】
----------------------------------------------------------
Workflow Description
Title: VACE 14B-Vensheng Video (Automatic Expansion Prompt Words)
Step 1, select the resolution ratio, enter the number 0~5
The number 0 represents custom, and the maximum value is recommended not to exceed 832
The number 1 represents the output ratio (1:1), a square screen, suitable for social media, highlighting the subject, and conducive to mobile browsing and sharing.
The number 2 represents the output ratio (3:4), a vertical rectangle, often used for portrait shooting and vertical videos, highlighting vertical content.
The number 3 represents the output ratio (4:3), a horizontal rectangle, suitable for early TV, surveillance videos and old movies, etc., to create a retro feel.
The number 4 represents the output ratio (9:16), an extra-long horizontal rectangle, which is the mainstream of TV, film and television and online videos, with a wide field of view and strong compatibility.
The number 5 represents the output ratio (16:9), an extra-long vertical rectangle, which fits the vertical screen mode of short video platforms such as Douyin, focusing on the subject of the character.
Step 2: Set the number of video frames. The default is 81 frames (5 seconds). The calculation method is 16 multiples + 1, so 4 seconds 65 frames, 5 seconds 81 frames, 6 seconds 97 frames, and so on.
Step 3: Enter text keywords, for example: beauty, 18 years old, close-up half-length photo, under a tree, with penetrating light spots on the face, Japanese style,
Step 4: Click Run. The test now uses the default parameters and can be completed in about 10 minutes. If the time exceeds too much, it is recommended to cancel the task and refresh the browser to run again.
----------------------------------------------------------
工作流内容介绍
标题:VACE 14B-文生视频(自动扩写提示词)
步骤一、选择分辨率比例,输入数字0~5
数字 0 代表自定义,建议最大值不要超过832数字 1 代表输出比例是(1 :1),正方形画面,适配社交媒体,突出主体,利于手机端浏览分享。数字 2 代表输出比例是(3 :4),竖版长方形,常用于人像拍摄与竖版视频,凸显垂直内容。数字 3 代表输出比例是(4 :3),横版长方形,适用于早期电视、监控视频和老电影等,可营造复古感。数字 4 代表输出比例是(9 :16),超长横版长方形,是电视、影视及网络视频主流,视野开阔,兼容性强。数字 5 代表输出比例是(16 :9),超长竖版长方形,契合抖音等短视频平台竖屏模式,聚焦人物主体。
步骤二、设置视频帧数,现默认是(5秒)81帧,计算方式是16的倍数+1,所以4秒65帧 5秒81帧 6秒97帧 以此类推
步骤三、输入文本关键词,例如:美女,18岁,近景半身照,树下,面部有穿透的光斑,日系小清新风格,
步骤四、点击运行即可,测试现在默认参数,10分钟左右可以跑完,如果时间超过太多建议取消任务,刷新浏览器后重新跑。
----------------------------------------------------------
RunningHub:Join Us to Claim Free Perks
Discussion
(No comments yet)
Loading...
Reviews
No reviews yet
Versions (1)
- latest (4 months ago)
Node Details
Primitive Nodes (24)
FluxResolutionNode (5)
InversionDemoLazyIndexSwitch (2)
JWInteger (1)
LoadWanVideoT5TextEncoder (1)
Note (1)
RH_Prompter (1)
Reroute (1)
WanVideoBlockSwap (1)
WanVideoDecode (1)
WanVideoEnhanceAVideo (1)
WanVideoExperimentalArgs (1)
WanVideoModelLoader (1)
WanVideoSLG (1)
WanVideoSampler (1)
WanVideoTeaCache (1)
WanVideoTextEncode (1)
WanVideoVACEEncode (1)
WanVideoVACEModelSelect (1)
WanVideoVAELoader (1)
Custom Nodes (24)
- easy showAnything (11)
- easy int (2)
- TextCombinerTwo (2)
- JjkText (3)
- Int (1)
- VHS_VideoCombine (1)
- MathExpression|pysssss (4)
Model Details
Checkpoints (0)
LoRAs (0)