logo
pub

Invoke 5.0 Update: New Canvas Features and Flux AI Integration

What's Happening with Invoke 5.0

Invoke AI has just launched a massive update—Invoke 5.0. This new release brings some mind-blowing features to the table, including a powerful Control Canvas, prompt templates, and full support for Flux AI models. It's a significant leap that aims to enhance your image creation experience tenfold.

Background on Invoke AI and Flux AI

Invoke AI has been a key player in the image generation space for a while. Their Canvas interface for Stable Diffusion was one of the first on the market. Flux AI, developed by Black Forest Labs, is an open-source image generation tool that excels at rendering accurate text and complex compositions. The integration of Flux AI models into Invoke 5.0 brings an extra layer of sophistication to the tool.

Impact and Reactions

There’s been a lot of buzz in the community about this update. With the new Control Canvas, users can now mix controlnets, IP adapters, and more with layers on an infinite canvas. This opens up endless creative possibilities. Partnering with Black Forest Labs to offer commercial-use licenses for Flux through their Professional Edition shows how serious Invoke is about supporting creative professionals.

Resources and Links

Step-by-Step Guide

Getting Started with Invoke 5.0

  1. Installation: Download the latest version from the GitHub page.
  2. Setup: Follow the installation instructions provided, ensuring you have the necessary dependencies installed. Ensure your invokeai.yaml file is properly configured, especially the ram and vram settings.
  3. Select a Model: Choose a Flux model that suits your needs. This can be done via the UI, where you will see options like Flux 'schnell', 'dev', and other models.
  4. Using Control Canvas:
    • Open the Control Canvas from the main menu.
    • Start by creating layers and adding elements like controlnets, IP adapters, and pixel-based tools.
    • Experiment with layering different elements to achieve complex compositions.
  5. Generate Images: Once your setup is ready, enter your text prompt, and let the AI generate images. You can use the prompt templates to save time.

Best Practices and Tips

  • Optimization: Ensure that your system’s RAM and VRAM are optimized in the ${INVOKEAI_ROOT}/invokeai.yaml configuration file.
  • Utilize Prompt Templates: Save, share, and reuse your favorite prompts to streamline your workflow.
  • Stay Updated: Regularly check for updates and join the community on Discord for support and new feature announcements.

Further Customizations

Invoke 5.0 offers numerous customization options:

  • Plugins and Extensions: Check the contributing guide to develop your own plugins.
  • Model Management: You can add models from various sources—URL, local path, or HuggingFace repo ID—and manage them efficiently using the model manager.
  • Node Extendibility: You can create custom nodes with specific parameters for personalized workflows.

Other User Needs

Users have expressed interest in various additional features:

  • Tablet Pen Pressure: While not supported in this release, it’s expected to be included in version 5.1.
  • Inpainting and Outpainting History: Features to keep and export the history are under development.
  • Support for GGUF: This is already in review and will be included in the next minor release.
  • LoRA Support: There are ongoing improvements to support more LoRA models and training variations.

Responding to Community Queries

  • Commercial Use of Flux: Clear guidelines are provided, and commercial licenses can be obtained through the Professional Edition of Invoke. Make sure to reach out if there’s any confusion regarding licensing terms.
  • Performance Issues: If you experience slow performance, it’s likely due to your system’s configuration. Check your RAM/VRAM settings and ensure they’re properly optimized.
  • AMD GPU Support: While Invoke predominantly supports Linux for AMD GPUs, users attempting to run it on Windows might face issues.

FAQs

How do I use the new Control Canvas?

The Control Canvas allows you to layer controlnets and other drawing tools for more complex compositions. You can start using it by selecting it from the main menu and experimenting with different settings.

Can I use Flux AI models commercially?

For non-commercial uses, you can use both Flux schnell and dev models. If you need to use them commercially, you'll have to obtain a commercial license from Black Forest Labs, available through Invoke's Professional Edition.

What are Prompt Templates and how do I use them?

Prompt Templates allow you to save, share, and reuse prompts. You can access this feature by navigating to the prompt template section in the UI and saving your favorite setups.

Why is generating images with Flux Dev so slow?

Flux Dev is more demanding on system resources. To mitigate this, ensure your RAM and VRAM configurations are properly set in the invokeai.yaml file.

Does Invoke 5.0 support my existing models and plugins?

Yes, you can install models from various sources (URL, local path, HuggingFace repo ID) and scan local folders for models. Plugin support is robust, and new plugins can be developed to extend the tool's functionality.

Is there support for tablet pen pressure?

Tablet pen pressure support isn't available in this release but is expected in the upcoming 5.1 release.

How can I optimize my system for better performance with Invoke 5.0?

Ensure that your ram and vram settings in the ${INVOKEAI_ROOT}/invokeai.yaml configuration file are optimized. Adjust these settings based on your system’s capacity to avoid unnecessary model offloads.

Can I keep and export the inpainting/outpainting history?

Currently, this feature is in development. Future updates will provide the ability to keep a history and export it for future use.

Invoke 5.0 is shaping up to be a game-changer with its new features and integrations. Dive in and start creating!

SHARE