> ## Documentation Index > Fetch the complete documentation index at: https://docs.flowscale.ai/llms.txt > Use this file to discover all available pages before exploring further. # Import/Create Workflows > Learn how to bring existing ComfyUI workflows into FlowScale AI or create new ones from scratch import ImportWorkflowsTroubleshooting from '/snippets/troubleshooting/import-workflows-basic.mdx'; Import your ComfyUI workflows into FlowScale AI for team collaboration and production deployment. This guide covers importing existing workflows and creating new ones in the cloud environment. **New to FlowScale AI?** You'll need to [create a project first](/getting-started/glossary/projects). Think of projects as collaborative workspaces where your team's workflows live and evolve together. ## Import Existing Workflows ### Step-by-Step Import Process In your FlowScale AI project dashboard, click the **"Import Workflow"** button Clean FlowScale project dashboard interface showing the main workspace with a prominently highlighted Import Workflow button in the top-right area. Shows project name, navigation sidebar, and empty workflow list state with clear call-to-action styling Select and upload your ComfyUI workflow JSON file (exported from ComfyUI using the "Save" option) Modal dialog overlay showing the workflow import interface with drag-and-drop file upload area, supported file format information (.json), and clear upload instructions. Includes progress indicator and file validation messaging FlowScale AI automatically detects required custom nodes and displays them for review Interface showing automated detection of custom nodes with a list view displaying node names, sources (GitHub URLs), and status indicators (detected/verified/missing). Includes resolution options and dependency management tools If conflicting nodes are detected, select the correct versions that your workflow actually uses Detailed conflict resolution interface showing side-by-side comparison of conflicting node versions with recommendation badges, compatibility notes, and user selection options. Includes version numbers and source information Click Confirm to add all necessary configuration for your workflow, and your imported project is ready to use! Step-by-step visual flow showing the entire import process from file upload through conflict resolution to successful integration, with progress indicators and completion confirmation ### Supported Import Format FlowScale AI imports **Workflow JSON** files from ComfyUI's "Save" function. **File Format Matters:** Only use JSON files exported directly from ComfyUI's Save function. Other export formats or manually edited files may cause import issues or lost functionality. ## Accessing ComfyUI Workspace Pods are cloud-hosted ComfyUI instances that provide on-demand access to GPU resources. ### Launch Process Click **"Start ComfyUI"** button in your project Project interface showing the prominent Start ComfyUI button with pod status indicators and workspace access controls. Shows resource availability and startup options Choose the pod configuration for your ComfyUI instance from the modal Modal dialog displaying available GPU pod options with detailed specifications (T4, A10G, L4, L40S, A100 40GB, A100 80GB, H100, H200, B200), pricing information, resource availability status, and performance recommendations for different workflow types While ComfyUI starts, custom nodes are installed automatically. View progress in System Logs Console/terminal interface showing real-time startup logs with custom node installation progress, dependency resolution status, and system initialization messages. Includes timestamps and log level indicators Once started, you'll see the ComfyUI workspace in CPU mode Full ComfyUI interface in CPU mode showing the node-based visual programming environment with sample workflow loaded, CPU mode indicator in top bar, and workspace tools/panels visible ### CPU vs GPU Mode **For workflow editing:** * Connecting nodes and adjusting parameters * Low cost for workflow development * Use for planning and structure building **For workflow execution:** * Running inference and testing outputs * Higher cost but enterprise GPU performance * Use only when generating results **Cost Optimization**: Always use CPU mode for editing workflows and only switch to GPU mode when you need to execute them. This prevents unnecessary credit burn. Close-up of the ComfyUI actions bar highlighting the CPU/GPU mode toggle switch with clear visual states, resource usage indicators, and cost implications display ## Model Management FlowScale AI provides centralized model management with public and private model libraries for team sharing. ### Model Setup Process Launch your ComfyUI instance in your project Run your workflow once to identify missing models (you'll see error messages) Error modal showing missing model notification with specific model names, suggested resolution actions, and links to model management tools. Clear error messaging with actionable next steps Look for required models in the dropdown menus of the workflow nodes themselves. If models are available in the public library, they will appear in the relevant node dropdowns within your workflow Models may appear under similar filenames—check carefully for variations or similar names. ComfyUI workflow node with expanded dropdown menu showing available models, model categories, and selection interface. Highlights newly added models and compatibility indicators If models aren't found publicly, navigate to **Private Models** section Dedicated models management page showing folder structure, uploaded models list, storage usage, and organization tools. Clean file browser-style interface with model metadata Create appropriate folder structure as required by ComfyUI or custom nodes Modal dialog for creating new model folders with folder type selection (checkpoints, loras, vae, embeddings), naming conventions, and organizational structure options Click **"Upload Model"** and either search for models or provide download links Tabbed interface showing multiple upload methods such as file upload, URL download, and model search. Includes progress tracking, validation, and metadata extraction features After model download completes, refresh ComfyUI from the actions bar Actions bar with highlighted refresh button for reloading models and nodes, showing refresh state indicators and workspace synchronization status Check the workflow node dropdowns again to confirm the newly added models are now available for selection ### Model Sources Pre-loaded models from Hugging Face, Civitai, and popular community sources Custom models, fine-tuned checkpoints, and specialized LoRAs Direct download from external sources using model URLs ### Folder Structure Guidelines When creating folders for models, follow ComfyUI conventions: * **Checkpoints**: `models/checkpoints/` * **LoRAs**: `models/loras/` * **VAE**: `models/vae/` * **Custom nodes**: Check the node's GitHub repository for specific folder requirements For custom node models, always refer to the node's GitHub repository README for correct folder naming conventions. ## Creating New Workflows Create workflows directly in FlowScale AI for immediate team collaboration and deployment readiness. ### Quick Start Method Click **"Create New Workflow"** button in your project Project dashboard with highlighted Create New Workflow button as alternative to importing, showing workflow templates and creation options A basic text-to-image workflow loads automatically in the ComfyUI workspace A basic text-to-image workflow loads automatically in the ComfyUI workspace Modify the workflow by adding, removing, or connecting nodes as needed Modify the workflow by adding, removing, or connecting nodes as needed Click **"Save"** in the actions bar and provide a descriptive commit message Click Save in the actions bar and provide a descriptive commit message ## Next Steps Transform workflows into production APIs and interactive interfaces