Wan 2.1 Fun Camera 14B
Generate high-quality videos with advanced camera control using the full 14B model
简介
The 'Wan 2.1 Fun Camera 14B' workflow in ComfyUI is designed to generate high-quality videos with advanced camera controls using the powerful Wan2.1 model. This workflow leverages a series of sophisticated nodes, including KSampler, CLIPTextEncode, and VAEDecode, to create visually compelling video outputs.
The process begins with loading necessary models through nodes like UNETLoader and VAELoader, which prepare the system for video generation. The workflow then uses the WanCameraImageToVideo and WanCameraEmbedding nodes to apply dynamic camera movements and settings, allowing for creative control over the video narrative.
Finally, the CreateVideo and SaveVideo nodes compile and store the generated video, ensuring a seamless production process.
Technically, this workflow is valuable because it integrates the Wan2.1 model with advanced diffusion techniques to produce videos that are not only high in quality but also rich in detail and motion. The use of CLIPVisionLoader and CLIPVisionEncode nodes ensures that the visual elements are aligned with the desired style and narrative, guided by the prompts provided. This makes the workflow particularly useful for creators looking to produce cinematic video content with precise control over visual aesthetics and camera dynamics.
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
- ComfyUI
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({
"3": {
"class_type": "KSampler",
"inputs": {
"cfg": 6,
"denoise": 1,
"latent_image": [
"56",
2
],
"model": [
"54",
0
],
"negative": [
"56",
1
],
"positive": [
"56",
0
],
"sampler_name": "uni_pc",
"scheduler": "simple",
"seed": 536456977716011,
"steps": 20
},
"_meta": {
"title": "KSampler"
}
},
"6": {
"class_type": "CLIPTextEncode",
"inputs": {
"clip": [
"38",
0
],
"text": "a vintage 1950s jazz bar, cinematic lighting, a well-dressed pianist playing passionately on an old upright piano, surrounded by saxophonists in dark suits, ambient smoke drifting in the warm air, smooth jazz atmosphere, camera slowly panning around the pianist, fingers moving on keys, soft spotlight from above, gentle camera shake and focus pull, close-up of piano keys, slow zoom in and out, realistic shadows and reflections, fun camera movement, Wan2.1 FunInP style, felt texture overlay, warm amber tones, retro film look, dynamic storytelling shot\n"
},
"_meta": {
"title": "CLIP Text Encode (Positive Prompt)"
}
},
"7": {
"class_type": "CLIPTextEncode",
"inputs": {
"clip": [
"38",
0
],
"text": "色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走"
},
"_meta": {
"title": "CLIP Text Encode (Negative Prompt)"
}
},
"8": {
"class_type": "VAEDecode",
"inputs": {
"samples": [
"3",
0
],
"vae": [
"39",
0
]
},
"_meta": {
"title": "VAE Decode"
}
},
"37": {
"class_type": "UNETLoader",
"inputs": {
"unet_name": "wan2.1_fun_camera_v1.1_14B_bf16.safetensors",
"weight_dtype": "default"
},
"_meta": {
"title": "Load Diffusion Model"
}
},
"38": {
"class_type": "CLIPLoader",
"inputs": {
"clip_name": "umt5_xxl_fp8_e4m3fn_scaled.safetensors",
"device": "default",
"type": "wan"
},
"_meta": {
"title": "Load CLIP"
}
},
"39": {
"class_type": "VAELoader",
"inputs": {
"vae_name": "wan_2.1_vae.safetensors"
},
"_meta": {
"title": "Load VAE"
}
},
"49": {
"class_type": "CLIPVisionLoader",
"inputs": {
"clip_name": "clip_vision_h.safetensors"
},
"_meta": {
"title": "Load CLIP Vision"
}
},
"51": {
"class_type": "CLIPVisionEncode",
"inputs": {
"clip_vision": [
"49",
0
],
"crop": "none",
"image": [
"52",
0
]
},
"_meta": {
"title": "CLIP Vision Encode"
}
},
"52": {
"class_type": "LoadImage",
"inputs": {
"image": "video_wan2.1_fun_camera_v1.1_14B_start_image.jpg"
},
"_meta": {
"title": "Load Image"
}
},
"54": {
"class_type": "ModelSamplingSD3",
"inputs": {
"model": [
"37",
0
],
"shift": 8
},
"_meta": {
"title": "ModelSamplingSD3"
}
},
"56": {
"class_type": "WanCameraImageToVideo",
"inputs": {
"batch_size": 1,
"camera_conditions": [
"57",
0
],
"clip_vision_output": [
"51",
0
],
"height": [
"57",
2
],
"length": [
"57",
3
],
"negative": [
"7",
0
],
"positive": [
"6",
0
],
"start_image": [
"52",
0
],
"vae": [
"39",
0
],
"width": [
"57",
1
]
},
"_meta": {
"title": "WanCameraImageToVideo"
}
},
"57": {
"class_type": "WanCameraEmbedding",
"inputs": {
"camera_pose": "Zoom Out",
"cx": 0.5,
"cy": 0.5,
"fx": 0.5,
"fy": 0.5,
"height": 512,
"length": 81,
"speed": 1,
"width": 512
},
"_meta": {
"title": "WanCameraEmbedding"
}
},
"58": {
"class_type": "CreateVideo",
"inputs": {
"bit_depth": "auto",
"codec": "none",
"color_space": "sRGB",
"fps": 16,
"images": [
"8",
0
]
},
"_meta": {
"title": "Create Video"
}
},
"59": {
"class_type": "SaveVideo",
"inputs": {
"filename_prefix": "video/ComfyUI",
"format": "auto",
"format.codec": "auto",
"video": [
"58",
0
]
},
"_meta": {
"title": "Save Video"
}
}
})
# 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 requires the Wan2.1 model and related components such as UNETLoader and VAELoader for video generation.
Yes, the WanCameraImageToVideo and WanCameraEmbedding nodes allow you to customize camera settings for dynamic video creation.
Use the CLIPTextEncode node to input detailed prompts that guide the visual style and narrative of the video.
The final video is saved in a standard video format, ensuring compatibility with most media players and platforms.