Dynamic 3D Logo Animations
Upload a vector image of your logo, and prompt your desired texture. Generate a textured 3D first and last frame with automated prompting for the final animation.
简介
The 'Dynamic 3D Logo Animations' workflow is designed to transform static vector images of logos into engaging 3D animations with textured surfaces. This workflow leverages a combination of advanced nodes and models to automate the creation of a visually appealing first and last frame, which are then used to generate a complete animation. Key nodes include the LoadImage node for importing your logo, the GeminiImage2Node for converting 2D images into 3D representations, and the ByteDanceFirstLastFrameNode, which intelligently generates the first and last frames of the animation.
The workflow also utilizes models like ByteDance and Google Gemini to ensure high-quality texture application and animation fluidity.
Technically, this workflow automates the traditionally complex process of 3D animation by employing AI-driven nodes that handle texture mapping and frame interpolation. The Seedream model is particularly effective in generating realistic textures, while the Nano Banana Pro model ensures smooth transitions between frames.
By automating these processes, the workflow allows users to focus on creative input, such as selecting the desired texture, while the system handles the technical intricacies of animation generation. This makes it an invaluable tool for brand designers looking to enhance their visual content with minimal effort.
你需要提供
- 正在读取工作流
在 ComfyUI 中运行。 此工作流无法在本页面运行。你可以在 Comfy Cloud 中打开（已预先加载所需内容），或下载后在自己的 ComfyUI 中使用。
在 Cloud 中试用查看工作流内部
节点图的只读预览。在 Comfy Cloud 中打开或下载后即可编辑节点。
Loading the graph
Open full-size workflow preview- 作者
- Rob
需要自定义节点
- 正在读取工作流
通过代码使用此工作流
尽可能根据工作流图和输入自动生成代码，否则显示已保存的示例。运行代码可能产生费用，请先核对凭据和端点。
What you need
- Runs on
- Comfy Cloud
- 密钥
- COMFY_API_KEY
需要付费的 Cloud 方案和额度。
# Python 3.10+: pip install comfy-sdk==0.4.0
import os
from comfy_sdk import Comfy
api_key = os.environ.get("COMFY_API_KEY", "").strip()
if not api_key: raise RuntimeError("Set COMFY_API_KEY")
os.environ.setdefault("COMFY_BASE_URL", "https://cloud.comfy.org")
client = Comfy(api_key=api_key)
workflow = client.workflows.from_json({
"1": {
"class_type": "LoadImage",
"inputs": {
"image": "comfy_logo_c.png"
},
"_meta": {
"title": "Load Logo Image"
}
},
"2": {
"class_type": "GeminiImage2Node",
"inputs": {
"aspect_ratio": "1:1",
"images": [
"1",
0
],
"model": "gemini-3-pro-image-preview",
"prompt": "Create a simple 3D version of this logo, it is colored white, slightly extruded, floating against a solid black background taken from a low left angle, giving it a heroic feel. Taken with a long focal length.\n\n",
"resolution": "1K",
"response_modalities": "IMAGE+TEXT",
"seed": 12345,
"system_prompt": "You are an expert image-generation engine. You must ALWAYS produce an image.\nInterpret all user input—regardless of format, intent, or abstraction—as literal visual directives for image composition.\nIf a prompt is conversational or lacks specific visual details, you must creatively invent a concrete visual scenario that depicts the concept.\nPrioritize generating the visual representation above any text, formatting, or conversational requests."
},
"_meta": {
"title": "Nano Banana Pro (Google Gemini Image)"
}
},
"3": {
"class_type": "SaveImage",
"inputs": {
"filename_prefix": "ComfyUI-3d_logo",
"images": [
"2",
0
]
},
"_meta": {
"title": "Save Image"
}
},
"4": {
"class_type": "GeminiImage2Node",
"inputs": {
"aspect_ratio": "1:1",
"images": [
"2",
0
],
"model": "gemini-3-pro-image-preview",
"prompt": "Generate this logo made out of glowing green ASCII code, maintain the depth and composition of the logo while adding the desired texture element",
"resolution": "1K",
"response_modalities": "IMAGE+TEXT",
"seed": 123,
"system_prompt": "You are an expert image-generation engine. You must ALWAYS produce an image.\nInterpret all user input—regardless of format, intent, or abstraction—as literal visual directives for image composition.\nIf a prompt is conversational or lacks specific visual details, you must creatively invent a concrete visual scenario that depicts the concept.\nPrioritize generating the visual representation above any text, formatting, or conversational requests."
},
"_meta": {
"title": "Nano Banana Pro (Google Gemini Image)"
}
},
"5": {
"class_type": "SaveImage",
"inputs": {
"filename_prefix": "ComfyUI-3d_logo_textured_1",
"images": [
"4",
0
]
},
"_meta": {
"title": "Save Image"
}
},
"6": {
"class_type": "GeminiImage2Node",
"inputs": {
"aspect_ratio": "1:1",
"images": [
"4",
0
],
"model": "gemini-3-pro-image-preview",
"prompt": "Generate this image 5 seconds into the future after much dynamic action, maintain the overall shape of the textured logo\n",
"resolution": "1K",
"response_modalities": "IMAGE+TEXT",
"seed": 321,
"system_prompt": "You are an expert image-generation engine. You must ALWAYS produce an image.\nInterpret all user input—regardless of format, intent, or abstraction—as literal visual directives for image composition.\nIf a prompt is conversational or lacks specific visual details, you must creatively invent a concrete visual scenario that depicts the concept.\nPrioritize generating the visual representation above any text, formatting, or conversational requests."
},
"_meta": {
"title": "Nano Banana Pro (Google Gemini Image)"
}
},
"7": {
"class_type": "SaveImage",
"inputs": {
"filename_prefix": "ComfyUI-3d_logo_textured_2",
"images": [
"6",
0
]
},
"_meta": {
"title": "Save Image"
}
},
"9": {
"class_type": "ByteDanceFirstLastFrameNode",
"inputs": {
"aspect_ratio": "adaptive",
"camera_fixed": False,
"duration": 5,
"first_frame": [
"4",
0
],
"generate_audio": False,
"last_frame": [
"6",
0
],
"model": "seedance-1-0-pro-250528",
"prompt": [
"22:17",
0
],
"resolution": "1080p",
"seed": 123,
"watermark": False
},
"_meta": {
"title": "ByteDance First-Last-Frame to Video"
}
},
"16": {
"class_type": "SaveVideo",
"inputs": {
"filename_prefix": "video/ComfyUI-logo_animation",
"format": "auto",
"format.codec": "auto",
"video": [
"9",
0
]
},
"_meta": {
"title": "Save Video"
}
},
"23": {
"class_type": "PreviewAny",
"inputs": {
"source": [
"22:17",
0
]
},
"_meta": {
"title": "Preview Transition Prompt"
}
},
"22:17": {
"class_type": "GeminiNode",
"inputs": {
"images": [
"22:20",
0
],
"model": "gemini-2.5-flash",
"prompt": "STEP 1: ANALYZE BOTH FRAMES\n\nFor each image, identify:\n\nVisual Elements:\nForm (shape, 3D, position)\nTexture (surface details, patterns)\nColor (hues, contrast, temperature)\nBackground (environment, depth)\nLighting (source, quality, direction, mood)\nPerspective (angle, focal point)\nMood (tone, energy, aesthetic)\n\nTransition Strategy:\nTransition order always goes from image 1 to image 2.\nShared elements connecting both frames\nChallenges (dimension shifts, style changes)\nOpportunities (natural bridges between frames)\n\n\nSTEP 2: BUILD THE PROMPT (≤50 WORDS)\nStructure:\n[Subject + Action] → [Beginning → Middle → End] + [MODIFIERS]\n\nRules:\nA. Lead with concrete subject + action verb\nThe transitions must have a lot of dynamic action.\nUse specific nouns and active verbs\nAvoid vague or passive language\n\nB. Map the journey in 3 beats\nDescribe HOW things transform, not just THAT they do\nStart state → transition → end state\n\nC. Use dynamic cinematic motion verbs\nGood: dissolves, spirals, cascades, erupts, morphs, fragments, reconverges\nAvoid: changes, moves, transforms (alone)\n\n\nD. State spatial transformations\nDimension changes: \"flat to volumetric\", \"2D to 3D\"\nScale shifts: \"zooms into detail\", \"pulls back to reveal\"\n\nE. Define visual style\nColors: literal names (teal, vermilion, neon green)\nLighting: volumetric, rim-lit, soft diffusion, bloom\nTexture: grain, scratches, chromatic aberration\nFPS: \"slow-motion\", \"cinematic 24fps\", \"digital rain FPS\"\n\nF. Add 1-3 bracketed modifiers\nFormat: [DESCRIPTOR IN CAPS]\nExamples: [DYNAMIC CAMERA PUSH], [VOLUMETRIC GLOW], [SLOW MOTION], [FAST PACED ENERGY]\n\n\nSTEP 3: OUTPUT\nONLY output the completed transition prompt. Nothing else.\n\nDo NOT include:\n\nProcess explanations\nStructural breakdowns\nAnalysis notes\nFollow-up questions",
"seed": 111,
"system_prompt": ""
},
"_meta": {
"title": "Google Gemini"
}
},
"22:20": {
"class_type": "ImageBatch",
"inputs": {
"image1": [
"6",
0
],
"image2": [
"4",
0
]
},
"_meta": {
"title": "Batch Images (DEPRECATED)"
}
}
})
# run() also passes api_key to the partner nodes in this workflow.
job = client.run(workflow, api_key=api_key)常见问答
The workflow is optimized for vector images, such as SVG, to ensure the best quality in 3D conversion.
Textures can be selected based on your brand's aesthetic. The Seedream model offers a variety of realistic options to choose from.
Ensure that the ByteDanceFirstLastFrameNode is correctly configured, as it handles the frame interpolation for smooth transitions.
Yes, the workflow is designed to be user-friendly, automating complex tasks while allowing for creative input.