Hello everyone! It's been half a year since our last update report, and this time we have a wealth of updates to share, including regular node and model updates as well as a series of important functional improvements and additions! Let's dive in!
# 📲 VApp Major Update: Support for Publishing as Paid Mini Programs!
Starting now, you can publish your VApp as a paid WeChat Mini Program, allowing you to share it with more people and earn revenue!
You can set free trial quotas and pricing for your VApp, and once it is published and approved, it will be available within the VectorVein WeChat Mini Program.
When users pay to use your VApp Mini Program, the payment amount will automatically be credited to your account balance, which you can view and manage at My Balance.
Try out this personal image transformation mini program:
# Other VApp Updates
- Brand customization (name, logo, page title)
- Support for managing multiple applications with a single key
- API call interface provided
- Data statistics visualization
Data Statistics
# 🖼️ Significant Enhancement in Image Processing Capabilities!
Image processing has been a key focus area for VectorVein recently, and this update brings several new image processing nodes to significantly enhance your creative capabilities!
# Core Image Processing Nodes
- 【Local Redrawing】: Supports both intelligent redrawing and custom redrawing, easily modify specific areas of an image, fix image flaws, and replace image elements.
- Workflow Template: Local Image Redrawing
Local Redrawing 
Local Redrawing
- 【Image Segmentation】: Accurately segment images using prompts or coordinate points, easily extract specific objects.
- 【PuLID】: While maintaining facial features, change the scene, pose, and expression through prompts.
# New Text-to-Image AI Models
【FLUX.1】: A 12B parameter open-source model developed by Black Forest Labs, capable of generating high-quality images in the style of Midjourney.
- The largest model within FLUX.1 is FLUX.1-Pro-Ultra, which can produce ultra-realistic images at 2k resolution. It supports native mode generation, making the generated images more closely resemble real photographs.
- Workflow Template: FLUX.1 Pro Ultra
FLUX.1
【Recraft】: Currently the top-ranked text-to-image model, supporting high-quality image and vector graphics generation, offering more possibilities for design scenarios.
- Workflow Template: Recraft Generating Digital Illustrations
- Recraft excels in generating digital illustrations and can produce SVG vector graphics, making it ideal for creating posters, illustrations, logos, etc.
Recraft 
Recraft
【Kolors】: A Chinese-characteristic text-to-image model developed by the Kolors team at Kuaishou, with outstanding Chinese language understanding capabilities.
Kolors
【Background Generation】: Optimized for e-commerce product images, this feature intelligently extends backgrounds to achieve natural light and shadow integration.
Background Generation
# Basic Image Tools
- [Image Background Removal]: Quick and precise image cutting
- [Image Editing]: Multi-functional image processing
- [Image Watermarking]: Batch add professional watermarks
👉 View Complete Image Generation Node Documentation
👉 View Complete Multimedia Editing Node Documentation
# 🎬 New Audio and Video Processing!
# Video Generation and Editing
- [CogVideoX]: A large video generation model developed by ZhiPu AI, generating videos with one click from text/images.
- [Keling Video Generation]: Supports high-quality longer duration videos and multiple aspect ratios.
- [Video Editing]: Supports basic functions such as trimming, rotating, and watermarking.
- [Video Screenshotting]: Supports taking screenshots at intervals or specific time points.
# Audio Processing Suite
- [Voice Cloning]: Replicate the voice characteristics you like
- [Sound Effect Generation]: Create unique sound effects through natural language descriptions
- [Audio Editing]: Supports functions such as trimming, volume adjustment, and speed control
👉 View Complete Video Generation Documentation
👉 View Complete Audio Generation Documentation
# ⚙️ New Control Nodes
- [Human Feedback]: Pause the workflow during execution to wait for human feedback
- [Loop Execution]: Loop through sub-workflows
- [Branch Workflow]: Smartly select and execute different sub-workflows based on conditions
👉 View Control Flow Node Documentation
# 🚀 API and Model Updates
# API Updates
- Upgraded to version 20240508
- Unified input parameters to
input_fields - Supports synchronous calls
- Simplified output results
# New Models
- OpenAI Model Series: gpt-4o, gpt-4o-mini, o1-preview, o1-mini
- Baichuan Intelligence Model Series: Baichuan4, Baichuan3
- ZeroOne Models Series: Yi-large, Yi-medium, Yi-spark, Yi-lightning
- Anthropic Model Series: claude-3-5-haiku
- x.ai Model Series: grok-beta
- Google Gemini Model Series: gemini-1.5-flash
- MiniMax Model Series: abab7-chat-preview
- Zhipu AI Model Series: glm-4-plus, glm-4-0520, glm-4-air, glm-4-airx, glm-4-flash
- Qwen2.5 Series of Tongyi Qianwen Models
# ✨ Other Functional Optimizations
- File reading node supports high-quality PDF parsing with OCR recognition, automatic layout, and table recognition capabilities
- Points balance warning
- PDF to image conversion support
- Mind map support for XMIND format
- Email support for URL type attachments
- VApp maximum points limit
- Global support for Latex rendering
- New interactive teaching examples
- Table output node
- Document node supports PDF output
- Automatic extraction of ZIP files
- Node deactivation feature
Thank you for your support! Join the beta testing group for more exchanges, add our enterprise WeChat:
