Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Update from feeds: https://galaxyproject.org/news/2024-11-26-flux/ #152

Conversation

galaxy-social-bot[bot]
Copy link
Contributor

This PR is created automatically by a feeds bot.
Update since 2024-11-14

Processed:
Text-to-Image Magic: FLUX Models Now in Galaxy!

Copy link
Contributor

👋 Hello! I'm your friendly social media assistant. Below are the previews of this post:
posts/feed_bot/galaxy_europe/2024-11-26-flux-group1.md

mastodon-eu-freiburg

📝 New blog post Released!
https://galaxyproject.org/news/2024-11-26-flux/

Using Black Forest Labs' FLUX Models Inside Galaxy: A Guide to Text-to-Image Integration

In previous news, we have written about the speech-to-text tool Whisper: https://galaxyproject.org/news/2024-04-25-whisper-tool/ and the ChatGPT tool: https://galaxyproject.org/news/2024-09-02-chat-gpt/ in Galaxy. Today, we would like to talk about a new tool that allows you to create images from text.
The Black Forest Labs (BFL): https://blackforestlabs.ai/ FLUX series has made significant strides in the world (1/9)

of generative AI, providing robust text-to-image capabilities
tailored for various use cases. At Freiburg University, we've integrated FLUX.1 [schnell] and FLUX.1 [dev] into the Galaxy platform, enabling researchers and developers to generate
high-quality images directly within Galaxy's user-friendly environment.

In this blog post, we'll explore the FLUX.1 models, their unique attributes, and how to use them effectively inside Galaxy.

Overview of FLUX Models

FLUX.1 [schnell]

  • (2/9)

Purpose: Designed for speed.

FLUX.1 [dev]

Both models provide easy access and (3/9)

versatility, making them excellent choices for text-to-image projects. While FLUX.1 [schnell] focuses on performance, FLUX.1 [dev] bridges the gap between quality and efficiency.

Why Use FLUX Models Inside Galaxy?

Galaxy's open platform supports workflows across life sciences, data analysis, and beyond. By integrating FLUX models, users can:

  • Generate visualizations: Convert textual descriptions into detailed images.
  • Streamline workflows: Incorporate generative AI into multi-step (4/9)

Galaxy pipelines.

  • Leverage open weights: FLUX's availability under open licenses aligns with Galaxy's ethos of accessibility and reproducibility.

To ensure high performance and speed, we are using a CUDA-based container to run the FLUX models on GPUs. This setup allows for fast inference times when working with the models from Hugging Face.
By leveraging Hugging Face’s pre-trained models, we can efficiently load the weights and run inference directly within the Galaxy environment. The (5/9)

CUDA-based container ensures that the heavy
computations involved in text-to-image generation are handled smoothly, providing users with a seamless experience.

Setting Up FLUX in Galaxy

The FLUX models are already available on UseGalaxy.eu.

Run FLUX on usegalaxy.eu!: https://usegalaxy.eu/root?tool_id=black_forest_labs_flux

Simply log in to your Galaxy instance and search for the FLUX tool in the tool panel.

Select the desired model (FLUX.1 [schnell] or FLUX.1 [dev]) based on your requirements.

Please be adviced (6/9)

that dev model can not be used for commercial use.

Choose between File or Text based input and select your txt file or enter your text prompt in the designated input field.
Click "Run Tool" and let Galaxy handle the rest!
Once the process is complete, you’ll find the generated image in your Galaxy history, ready to view, share, or use in your workflows.

For example, I used the text prompt "A galaxy full of stars" to generate an image. Here's the result history: https://usegalaxy.eu/u/brilliantarash/h/flux: (7/9)

Text Prompt: A galaxy full of stars

Generated Image: (8/9)

You can try out your own prompts and see what unique images FLUX can generate!

@[email protected]
#UseGalaxy #GalaxyProject #UniFreiburg #EOSC #EuroScienceGateway (9/9)

FLUX_on_text_prompt_(A_galaxy_full_of_stars)

matrix-eu-announce

FLUX_on_text_prompt_(A_galaxy_full_of_stars)

📝 New blog post Released! https://galaxyproject.org/news/2024-11-26-flux/

Using Black Forest Labs' FLUX Models Inside Galaxy: A Guide to Text-to-Image Integration

In previous news, we have written about the speech-to-text tool Whisper and the ChatGPT tool in Galaxy. Today, we would like to talk about a new tool that allows you to create images from text. The Black Forest Labs (BFL) FLUX series has made significant strides in the world of generative AI, providing robust text-to-image capabilities tailored for various use cases. At Freiburg University, we've integrated FLUX.1 [schnell] and FLUX.1 [dev] into the Galaxy platform, enabling researchers and developers to generate high-quality images directly within Galaxy's user-friendly environment.

In this blog post, we'll explore the FLUX.1 models, their unique attributes, and how to use them effectively inside Galaxy.

Overview of FLUX Models

FLUX.1 [schnell]

  • Purpose: Designed for speed.
  • License: Openly available under the Apache 2.0 license.
  • Use Case: Ideal for personal use and rapid prototyping.

FLUX.1 [dev]

  • Purpose: Guidance-distilled model for non-commercial applications, offering near-professional quality.
  • License: Open-weight model intended for non-commercial contexts.
  • Use Case: Suited for research and non-commercial exploration.

Both models provide easy access and versatility, making them excellent choices for text-to-image projects. While FLUX.1 [schnell] focuses on performance, FLUX.1 [dev] bridges the gap between quality and efficiency.

Why Use FLUX Models Inside Galaxy?

Galaxy's open platform supports workflows across life sciences, data analysis, and beyond. By integrating FLUX models, users can:

  • Generate visualizations: Convert textual descriptions into detailed images.
  • Streamline workflows: Incorporate generative AI into multi-step Galaxy pipelines.
  • Leverage open weights: FLUX's availability under open licenses aligns with Galaxy's ethos of accessibility and reproducibility.

To ensure high performance and speed, we are using a CUDA-based container to run the FLUX models on GPUs. This setup allows for fast inference times when working with the models from Hugging Face. By leveraging Hugging Face’s pre-trained models, we can efficiently load the weights and run inference directly within the Galaxy environment. The CUDA-based container ensures that the heavy computations involved in text-to-image generation are handled smoothly, providing users with a seamless experience.

Setting Up FLUX in Galaxy

The FLUX models are already available on UseGalaxy.eu.

Run FLUX on usegalaxy.eu!

Simply log in to your Galaxy instance and search for the FLUX tool in the tool panel.

Select the desired model (FLUX.1 [schnell] or FLUX.1 [dev]) based on your requirements.

Please be adviced that dev model can not be used for commercial use.

Choose between File or Text based input and select your txt file or enter your text prompt in the designated input field. Click "Run Tool" and let Galaxy handle the rest! Once the process is complete, you’ll find the generated image in your Galaxy history, ready to view, share, or use in your workflows.

For example, I used the text prompt "A galaxy full of stars" to generate an image. Here's the result history:

Text Prompt: A galaxy full of stars

Generated Image:

You can try out your own prompts and see what unique images FLUX can generate!

linkedin-galaxyproject

📝 New blog post Released!
https://galaxyproject.org/news/2024-11-26-flux/

Using Black Forest Labs' FLUX Models Inside Galaxy: A Guide to Text-to-Image Integration

In previous news, we have written about the speech-to-text tool Whisper: https://galaxyproject.org/news/2024-04-25-whisper-tool/ and the ChatGPT tool: https://galaxyproject.org/news/2024-09-02-chat-gpt/ in Galaxy. Today, we would like to talk about a new tool that allows you to create images from text.
The Black Forest Labs (BFL): https://blackforestlabs.ai/ FLUX series has made significant strides in the world of generative AI, providing robust text-to-image capabilities
tailored for various use cases. At Freiburg University, we've integrated FLUX.1 [schnell] and FLUX.1 [dev] into the Galaxy platform, enabling researchers and developers to generate
high-quality images directly within Galaxy's user-friendly environment.

In this blog post, we'll explore the FLUX.1 models, their unique attributes, and how to use them effectively inside Galaxy.

Overview of FLUX Models

FLUX.1 [schnell]

FLUX.1 [dev]

Both models provide easy access and versatility, making them excellent choices for text-to-image projects. While FLUX.1 [schnell] focuses on performance, FLUX.1 [dev] bridges the gap between quality and efficiency.

Why Use FLUX Models Inside Galaxy?

Galaxy's open platform supports workflows across life sciences, data analysis, and beyond. By integrating FLUX models, users can:

  • Generate visualizations: Convert textual descriptions into detailed images.
  • Streamline workflows: Incorporate generative AI into multi-step Galaxy pipelines.
  • Leverage open weights: FLUX's availability under open licenses aligns with Galaxy's ethos of accessibility and reproducibility.

To ensure high performance and speed, we are using a CUDA-based container to run the FLUX models on GPUs. This setup allows for fast inference times when working with the models from Hugging Face.
By leveraging Hugging Face’s pre-trained models, we can efficiently load the weights and run inference directly within the Galaxy environment. The CUDA-based container ensures that the heavy
computations involved in text-to-image generation are handled smoothly, providing users with a seamless experience.

Setting Up FLUX in Galaxy

The FLUX models are already available on UseGalaxy.eu.

Run FLUX on usegalaxy.eu!: https://usegalaxy.eu/root?tool_id=black_forest_labs_flux (1/3)

Simply log in to your Galaxy instance and search for the FLUX tool in the tool panel.

Select the desired model (FLUX.1 [schnell] or FLUX.1 [dev]) based on your requirements.

Please be adviced that dev model can not be used for commercial use.

Choose between File or Text based input and select your txt file or enter your text prompt in the designated input field.
Click "Run Tool" and let Galaxy handle the rest!
Once the process is complete, you’ll find the generated image in your Galaxy history, ready to view, share, or use in your workflows.

For example, I used the text prompt "A galaxy full of stars" to generate an image. Here's the result history: https://usegalaxy.eu/u/brilliantarash/h/flux:

Text Prompt: A galaxy full of stars

Generated Image: (2/3)

You can try out your own prompts and see what unique images FLUX can generate!

#UseGalaxy #GalaxyProject #UniFreiburg #EOSC #EuroScienceGateway (3/3)

FLUX_on_text_prompt_(A_galaxy_full_of_stars)

Copy link
Contributor

👋 Hello! I'm your friendly social media assistant. Below are the previews of this post:
posts/feed_bot/galaxy_europe/2024-11-26-flux-group2.md

bluesky-galaxyproject

📝 New blog post Released!
https://galaxyproject.org/news/2024-11-26-flux/

Using Black Forest Labs' FLUX Models Inside Galaxy: A Guide to Text-to-Image Integration

In previous news, we have written about the speech-to-text tool Whisper: (1/15)

https://galaxyproject.org/news/2024-04-25-whisper-tool/ and the ChatGPT tool: https://galaxyproject.org/news/2024-09-02-chat-gpt/ in Galaxy. Today, we would like to talk about a new tool that allows you to create images from text.
The Black Forest Labs (BFL): https://blackforestlabs.ai/ FLUX (2/15)

series has made significant strides in the world of generative AI, providing robust text-to-image capabilities
tailored for various use cases. At Freiburg University, we've integrated FLUX.1 [schnell] and FLUX.1 [dev] into the Galaxy platform, enabling researchers and developers to generate (3/15)

high-quality images directly within Galaxy's user-friendly environment.

In this blog post, we'll explore the FLUX.1 models, their unique attributes, and how to use them effectively inside Galaxy.

Overview of FLUX Models

FLUX.1 [schnell]

  • Purpose: Designed for speed.
  • License: (4/15)

https://github.com/black-forest-labs/flux/blob/main/model_licenses/LICENSE-FLUX1-schnell: Openly available under the Apache 2.0 license.

  • Use Case: Ideal for personal use and rapid prototyping.

FLUX.1 [dev]

  • Purpose: Guidance-distilled model for non-commercial applications, offering (5/15)

near-professional quality.

Both models provide easy access and (6/15)

versatility, making them excellent choices for text-to-image projects. While FLUX.1 [schnell] focuses on performance, FLUX.1 [dev] bridges the gap between quality and efficiency.

Why Use FLUX Models Inside Galaxy?

Galaxy's open platform supports workflows across life sciences, data (7/15)

analysis, and beyond. By integrating FLUX models, users can:

  • Generate visualizations: Convert textual descriptions into detailed images.
  • Streamline workflows: Incorporate generative AI into multi-step Galaxy pipelines.
  • Leverage open weights: FLUX's availability under open licenses (8/15)

aligns with Galaxy's ethos of accessibility and reproducibility.

To ensure high performance and speed, we are using a CUDA-based container to run the FLUX models on GPUs. This setup allows for fast inference times when working with the models from Hugging Face.
By leveraging Hugging Face’s (9/15)

pre-trained models, we can efficiently load the weights and run inference directly within the Galaxy environment. The CUDA-based container ensures that the heavy
computations involved in text-to-image generation are handled smoothly, providing users with a seamless experience.

Setting Up (10/15)

FLUX in Galaxy

The FLUX models are already available on UseGalaxy.eu.

Run FLUX on usegalaxy.eu!: https://usegalaxy.eu/root?tool_id=black_forest_labs_flux

Simply log in to your Galaxy instance and search for the FLUX tool in the tool panel.

Select the desired model (FLUX.1 [schnell] or (11/15)

FLUX.1 [dev]) based on your requirements.

Please be adviced that dev model can not be used for commercial use.

Choose between File or Text based input and select your txt file or enter your text prompt in the designated input field.
Click "Run Tool" and let Galaxy handle the rest!
Once the (12/15)

process is complete, you’ll find the generated image in your Galaxy history, ready to view, share, or use in your workflows.

For example, I used the text prompt "A galaxy full of stars" to generate an image. Here's the result history: https://usegalaxy.eu/u/brilliantarash/h/flux:

Text (13/15)

Prompt: A galaxy full of stars

Generated Image: (14/15)

You can try out your own prompts and see what unique images FLUX can generate!

@galaxyproject.bsky.social
#UseGalaxy #GalaxyProject #UniFreiburg #EOSC #EuroScienceGateway (15/15)

FLUX_on_text_prompt_(A_galaxy_full_of_stars)

@bgruening bgruening closed this Dec 17, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant