No-code platform to build and deploy web apps and APIs using ComfyUI workflows. Allows sharing workflows, scaling AI operations, and customizing apps without complex setups.
Features
No-Code ComfyUI Developer Cloud
Allows users to turn ComfyUI workflows into web apps and APIs without writing code, enabling easy development, deployment, and sharing of workflows.
Upload and Run Workflows
Users can upload their workflows, choose parameters, and run apps on scalable cloud infrastructure with minimal setup.
Workflow Sharing
Enables sharing workflows with others, even if they don't have ComfyUI, via easy-to-use web apps or custom API endpoints.
Cloud-Based AI Operations
Run apps and APIs on powerful cloud infrastructure instantly, providing scalable solutions for AI operations.
Open Source Customization
Users can download the source code for ViewComfy web apps, allowing for personalized customization.
One click API deployment
Deploy APIs with a single click on any hardware you choose, simplifying the process of managing and deploying APIs.
No code app builder
Create beautiful user interfaces for your workflows without writing code, enabling quick and easy visualization of your processes.
Optimized solution with minimal cold starts
Ensure your applications run efficiently with minimal delays during startup, providing an optimized and seamless experience.
HunyanVideo LoRA Select
Allows you to choose the LoRA and set its strength, which influences the output during video generation. Setting the strength to 0 ignores it entirely, while setting to 1 maximizes its effect.
HunyanVideo Model Loader
Node responsible for loading the Hunyan video model. Offers a b16 and a b9e model, with the latter using less GPU space but offering lower output quality.
HunyanVideo TextEncode
Lets you set the prompt for video creation using a specific format for subject, action, scene, style, and quality description.
HunyanVideo Sampler
Node where video generation occurs, enabling you to specify width, height, and the number of frames. Provides recommended settings for Hunyan video based on resolution and aspect ratio.
Workflow Deployment
Use ViewComfy Cloud to deploy ComfyUI workflows as an API or a web app, simplifying the process without coding and managing dependencies. Scale across multiple GPUs.
Model Integration
Access a large library of models that can be linked to your deployment. The platform guides you through downloading models if they are not in the library.
UI Customization
Customize the user interface, group parameters by categories, add preview images, and manage output types to tailor the interface to your needs.
Deployment Setup
Select hardware type and number of GPUs for deployment, optimizing the setup for specific workflows. Automatically scales GPU usage based on demand.
Workflow Management
Manage deployments through the ViewComfy dashboard, install new custom nodes, and edit workflows, offering flexibility and efficient handling.
Stable Diffusion 3.5 Integration
Allows users to generate images using Stable Diffusion 3.5 in ComfyUI. Users need to download the sd3.5 files and add Stable Diffusion 3.5 or 'sd_xl_base_1.0' to their models folder for enabling this feature.
Load Checkpoint
Enables users to load specific checkpoints for the Stable Diffusion 3.5 model within ComfyUI. Users can load checkpoints to make customizations on the generated images.
EmptySDXLLatentImage
Provides settings for adjusting image dimensions, specifically width and height for better performance with Stable Diffusion 3.5. Suggested dimensions include multiples of 8, like 1024x1024.
ConditioningOffsetNoiseOffset
Allows fine-tuning of image generation by setting a noise offset parameter. Suggests using this with a value close to 1 for improved results.
CLIP Text Encoder
Enables users to input both positive and negative prompts to manipulate image generation.
Load Checkpoint
Selects the Stable Diffusion SDXL model used for upscaling. Flexibility to use any SDXL model offers broad compatibility.
Load LoRA
Allows the addition of LoRA to add new details to the image during generation. Multiple LoRAs can be stacked for customized results.
CLIP Text Encoder (Prompt)
Includes nodes for both positive and negative prompts, enhancing the upscaling process by improving image description and results.
Load Image
Uploads the image intended for upscaling, serving as the initial input for the process.
Ultimate SD Upscale
Core upscaling node that adjusts image size using a controlled scale factor. Works optimally with values between 2 and 4, with typical results between 0.3 and 0.6.
ComfyUI Integration
Allows users to create custom image-to-image transformations without writing code by using a visual node-based editor to set up workflows.
Jupyter Lab Environment
Sets up a Jupyter Lab environment to manage and interact with ComfyUI workflows directly from a web-based interface.
Pod Deployment
Enables users to select specific hardware configurations through a pod deployment system for the application setup, tailored to their processing needs.
Parameter Customization
Provides tools for setting starting images and parameters for transformation, allowing for fine-tuned results.