Freeway ML Feature Updates, October 27th

We want to thank all the users that reached out to provide feedback. Thanks to your valuable insights, we've squashed several bugs improving inpainting results, resolved some issues in the Recent View for users with many images, fixed downloads for images with long prompts, and more. Please keep the feedback coming!


Table of Contents

🖼 Image Generation Quality Improvement

We've made some updates to our AI models, and we're excited to share a significant improvement in image results. Image Operations now provide more accurate and realistic image results. Let's walk through a few examples.

Improvement: Wine is poured inside of the glass, previously it was outside the rim

Improvement: Fixed the base of the cup and better lighting

Improvement: More accurate prompt with hanging street lights

👌 Fine Details Quality Improvement

They say that the eyes are the window to the soul. Historically, AI image generation for eyes and other fine details has sometimes struggled with details such as eyes, resulting in output that can seem uncanny. Humans are incredibly perceptive to these details, especially when an image is a person. The Freeway team has updated our AI engine to improve generation of fine details such as eye pupils. Check it out:

Improvement: Eyes and hair

Close up of eye

✍️ Masked Style Swap

Last week we announced image inpainting enabling users to take their creativity to the next level. This week, we updated masked image styling operations to have significantly improved results. Here are some style transformations with the recent update. Here are some style transformations with the recent update.

Mask on face, hair and background remain the same

Mask on astronaut, asteroid, planets and rainbow remain the same

That's all for this week. Keep on creating!

AI Image Generator – create realistic images from text

Start creating for free

Keep reading

View all posts

10x Your Productivity with AI✨

Get started with Freeway's AI Agents today

Sign up for free