I keep hearing the question: “Will AI replace Photoshop?”
Well, I just saw an awesome post from Tianyu Xu where he puts Google’s latest image model, Gemini 2.5 Flash, to the test. He focused on the most hyped features, editing and blending, to see if it lives up to the hype.
Here’s the breakdown of what he discovered:
✨ The Good Stuff ✨
According to his tests, Gemini 2.5 Flash is a major upgrade. It’s great at:
- ✅ Keeping characters highly consistent across images.
- ✅ Replicating elements from your input images.
- ✅ Showing the same object from different angles.
- ✅ Producing higher-quality images than its predecessor.
He also noted it’s fast and super easy to use, which is a huge plus!
⚙️ The Hiccups ⚙️
It’s not perfect, though. Tianyu found a few weaknesses:
- ❌ You can’t control the aspect ratio.
- ❌ The image editing can sometimes look a bit like a “copy and paste” job.
- ❌ Character resolution can drop after an edit.
- ❌ It struggles to display long lines of text accurately in an image.
✍️ The Final Verdict
So, is it a Photoshop killer? Tianyu says not yet.
BUT, he found it works incredibly well for concept testing, building AI workflows, and creating social media content fast.
🚀 Pro-Tips from Tianyu 🚀
He shared a few tips to save you time:
- 💡 Use close-up shots of your characters or objects.
- 💡 Avoid creating images with a lot of text.
- 💡 If you get stuck, just start a new chat.
This is a game-changer for quick creative tasks! For the full deep-dive and to see the video he made, you’ve got to check out Tianyu Xu’s original post.