Release Notes

• Seedream 5 Lite – Edit Image Support:

Edit and refine images with Seedream 5 Lite using prompts and image inputs—making it easier to adjust visuals, explore variations, and improve outputs without starting from scratch.

• Run Flow Improvements:

Run Flow has been improved for smoother execution and a more reliable experience across operations.

• Media Inspector Enhancements:

Media Inspector now includes improved Details handling and Clear All support for a cleaner and more streamlined workflow.

• Layer Compositor Improvements:

Layer Compositor now supports zoom controls, better highlighting for items outside the canvas, and HTML Export support for improved usability and output flexibility.

• Huge File Upload Improvements:

Improved handling for large file uploads to provide a smoother and more stable upload experience.

• Motion Graphics Stability Fixes:

Optimizations to improve Motion Graphics stability.

• AI Video Vision

Analyze videos with AI to understand sentiment, mood, tone, and overall impact—just upload a video and let the model break it down for you.

Learn More

• Seedream 5.0 Lite Image Generation

Generate clean, high-quality images from simple text prompts with better consistency and faster results—great for quick ideas and visuals.

Learn More

• Edit Video with Seedance 2.0

Edit and refine your videos with more control—adjust style, improve consistency, and get smoother, more polished outputs.

Learn More

• Lock the Node

Keep important nodes exactly where they belong. Lock a node to prevent accidental edits or movement, making it easier to protect finalized steps in your flow.

Learn More

• Sticky Notes

Keep ideas, reminders, and feedback right where you need them. Add sticky notes to your canvas to capture thoughts, leave comments, and organize your flow visually.

Learn More

• Seedance 2 Video Generation

Bring ideas to life with our new Seedance 2 model. Create smoother, sharper, and more polished AI videos with better motion and cleaner visuals from a simple prompt.

Learn More

• Dark Theme
  
  Like working in the dark? Fluidfit now comes with Dark Theme. It follows your system setting by default, so it automatically matches the way you prefer to work.
  
  Learn More
  
  

• Motion Graphics
  
  Turn your extracted PSD layers into animated videos with Motion Graphics. Add movement to each layer and turn static designs into engaging visuals — all without leaving Fluidfit.
  
  Learn More

• Layer Splitter
  
  Got a layered Photoshop file? Layer Splitter pulls it apart for you. It extracts images and text from your PSD and turns them into ready-to-use layers in Fluidfit — so you can jump straight into building.
  
  Learn More
  
  

• Stability Improvements

  • • Run Flow API now supports Text and Audio nodes as outputs

    • General bug fixes and improvements in the sidebar UI

    • Fixed responsiveness issue in Prompt Editor where “Save” and “Cancel” buttons were hidden on certain screen sizes

    • Layer Compositor editor: Selecting an item in the preview now highlights the corresponding layer. Editing mode is activated only when the layer is explicitly selected.

    • Layer Compositor editor: Opacity control updated from input field to a slider for easier adjustments

• Generate Audio
  
  Fluidfit now lets you generate audio right inside your workflows. Just add a Generate Audio node, provide the audio script, and produce audio seamlessly as part of your flow. Your generated audio is ready to preview and use in downstream steps—quick, simple, and fun. 🎧
  
  Learn More

• Layer Compositor
  
  Want layered Photoshop files? You got it. With Layer Compositor, you can now combine images and text in Fluidfit and generate a fully layered PSD. It’s an easy way to build structured designs.
  
  Learn More
  
  

• Cleaner Sidebar Layout
  
  The sidebar is now reorganized with Media, Tools, and Image Editing grouped neatly — making everything easier to find and faster to use.

Learn More

• Prompt Editor Improvements
  
  Editing prompts just got clearer. The Prompt Editor now includes Save and Cancel buttons. If you close the dialog without saving, your changes are dismissed—no surprises.

Support for New AI Models

Fluidfit just got a little more powerful—more models, more possibilities. Now with Google Nano Banana 2 for image workflows and Kling 3.0 & Higgsfield for video, giving you more flexibility to build the way you want.

Learn More

AI Vision via API

You can now use AI Vision via the Fluidfit API—not just in the visual builder.

Send images, add prompts, and get structured results back (JSON or text) to plug directly into your apps, workflows, or automations.

Learn More

Stability Improvements

• AI Vision is now more resilient—it won’t fail even if a Media node is mistakenly connected as output.

• Issues with PDF Upload and Add Accessory features have now been resolved.

AI Vision

Introducing the AI Vision node in Fluidfit, which enables you to analyze images using advanced AI vision models and extract meaningful insights such as detected text, brands, descriptions, keywords, and product categories.

Just a quick note: AI Vision produces text output, so there’s no need to connect a Media node after it.

Learn More

Stability Improvements

• • We’ve implemented major infrastructure improvements to enhance system reliability and eliminate intermittent silent feature failures.

VerifAI – UX Improvements

• Enhanced Output Visibility: Clear visual differentiation between passed and failed cases using distinct color highlights and responsive bounding boxes.

• Improved Search Experience: New search interface enables filtering reports by validation status, rule, or reasoning.

• Optimized Results View: Increased number of results displayed per screen for improved readability and faster interpretation.

• Integrated Checklist View: Users can now view the checklist on the output screen for quicker reference and validation.
  
  Learn more about VerifAI 

Stability Improvements

• Flow Name Display: Flow name is now prominently visible within the Share Flow interface.

• Improved Layout: Shared user avatars are now repositioned to the flow thumbnail, providing more space for clearer flow name visibility.

Reshare flows

Fluidfit.ai now allows users to easily reshare flows with other users. Editors can share flows with either Editor or Viewer permissions, while Viewers can reshare flows with Viewer access only.
Learn More

Platform Architecture Upgrade for Google AI Models

We upgraded the underlying platform architecture to enhance performance, reliability, and scalability for Google AI model integrations.

The following models benefit from this improvement:

• Generate Image: Nano Banana (incl. Pro), Google Imagen 4 (Ultra & Fast)

• Edit Image: Nano Banana (incl. Pro)

• Generate Video: Veo 2, Veo 3.1 (incl. Fast) 

No functional changes to user workflows.

Stability Fixes

• • Credit Validation Before Feature Execution

Share flows with organization

Fluidfit.ai enables users to easily share flows with members of their organization, giving teams seamless access and collaboration. Flow owners can also manage or revoke access at any time using the Stop Sharing control, ensuring full control over who can view or edit a flow.

Learn More

Share and Collaborate on Flows in Fluidfit.ai

Fluidfit.ai makes it easy to collaborate on flows while maintaining clear ownership, controlled access, and full traceability. Flow sharing allows you to invite teammates or external collaborators, assign them specific roles, and work together securely—without losing control of your work.
Learn More

Vector Image Generation from Text or Images
Fluidfit.ai now supports vector image creation using two powerful tools:  Generate Vector and Vectorize Image. 
Learn More

Stability Improvements & Bug Fixes

• Admins can now reduce group credits and quickly transfer credits between groups within an organization.

• Group deletion should send back the unused credits to the org back.

Stability Improvements & Bug Fixes

This update focuses on improving the stability of the Create Mask tool:

• Composite Mask selections are now preserved after execution.

• Mask output downloads from Fluidfit app now follow a consistent naming convention:
  <input_file_name>_<prompt>_<number>

Support for New AI Models

Fluidfit now supports additional AI models across Generate Image, Generate Video, and AI Image Edit features. These new models expand the capabilities available in your flows and allow you to choose the most suitable model based on your use case.
Learn More

Stability Improvements & Bug Fixes

• Removed unnecessary pricing display for multi-model features.

Create Mask API
The Create Mask API now follows a standardized naming convention for mask outputs. Mask files are generated in the format: <filename>_<prompt>_<number>.png

Example:
TestFile_chair_1.png
TestFile_chair_2.png

Stability Improvements & Bug Fixes

• Media Inspector:
  Media inspector was not functional earlier in the flow refresh case, now fixed.

Create Mask
We have updated our Create Mask tool with new capabilities.
This tool now allows you to automatically isolate any object in an image by describing it in plain text. Fluidfit’s AI detects the subject and generates accurate masks in seconds, helping you work faster with full control over each element.
You can also invert masks or combine multiple masks into a single mask—all without manual drawing.
For more details , please refer:  Create Masks 

Stability Improvements & Bug Fixes

• Adjustments Feature:
  Removed the deprecated Adjustments feature to ensure a cleaner user experience.

• AI Edit – Reference Image Accuracy:
  Improved the accuracy of AI Edit outputs when using a reference image.

• Credit Deduction for Video Tools:
  Fixed an issue in the credit deduction flow affecting video-related tools.

User Credit Limit
Admins and Group Admins can now set a credit usage limit for users within their organization. A credit limit defines the maximum number of credits a user can consume. Once the limit is reached, the user will no longer be able to access credit-consuming features.
For more details , please refer:  User Credit Limit 

Stability Improvements & Bug Fixes

• Notification Emails (Multiple Admin Scenario):
  Fixed an issue where low credit notification emails were not sent to all administrators. Notifications are now delivered to all admins as expected.

• Prompt Editor Cleanup:
  Removed negative prompts from the Conditional, Mask, and Inpaint feature nodes.

Enhancement

• The Enhancement tool can be used as a component within Fluidfit flows to optimize images at any stage of processing. It provides fine-grained controls that allow users to adjust visual attributes such as brightness, contrast, color balance, sharpness, and tonal ranges, enabling best-in-class image output quality.

Reset to Default
Previously, some feature and tool option editors did not include a Reset to Default option. This functionality is now available, allowing you to quickly restore all values and the default model with a single click.

• The following features currently support Reset to Default:

• Enhancement

• Adjustments

• Generate Image

• File Format

• AI Add Accessory

• Generate 3D

• Rotate

• AI Image Edit

• Generate Video

• Resize

• Dress Change

• Create Video Preview

• Crop Image

• Image Overlay

• Remove Video Background

• AI Edit Video

• Add Text

• Upscale Video

• Add Background

• Inpaint

• Reframe Video

• Remove Background

• Upscale

• Mask

Hot Folder

Hot Folder enables automated execution of Fluidfit flows by monitoring an input folder. Users can trigger configured flows by placing input images into the designated folder and retrieve the processed output images from the corresponding download link.

Note:
This feature is disabled by default for all users.
To enable Hot Folder for your account, please contact your account manager with your account details.

Admin – Invite User Flow

Administrators can now invite new users to their organization directly from the Admin Panel. Invited users receive an email to reset their password and complete account setup before accessing the platform.

Notifications

Administrators can configure email notifications based on credit consumption limits. This helps teams proactively manage usage and avoid unexpected service interruptions.

Stronger Password Policy for New Users

New users are now required to follow an enhanced password policy for improved security. Passwords must meet the following criteria:

• Minimum of 8 characters

• At least one uppercase letter

• At least one lowercase letter

• At least one numeric character

• At least one special character
  
  

Improved Empty State Screens

Empty screens across the application have been redesigned to provide clearer guidance and a more user-friendly experience, improving overall usability and navigation.

Stability Improvements & Bug Fixes

The following stability issues have been addressed in this release:

• Failure messages during Template Copy operations

• Missing prompts when copying templates

• Missing default model configuration on feature nodes

• Errors encountered while generating images and videos with certain models

• Errors in image editing and accessory addition flows with specific models