The 5 AI image generation tools that we will compare

This page has 0 threads | Add post

Blogs

Finding our blogs useful? Please consider helping us to share more!

Comparing the 5 leading AI tools for image generaton from text Part two of an eight-part series of blogs
Having recently compared the 4 main AI tools for text prompts, I thought I'd do the same for image generation tools. In this blog series we compare Dall-E (via ChatGPT and separately via Copilot), Firefly, Midjourney and Stable Diffusion for 3 pre-defined tests to see which ones score most highly for cost, ease-of-use, speed, editing ability and above all for quality of image. A deep-dive review of text-to-image AI tools The 5 AI image generation tools that we will compare (this blog) The 3 pre-defined tests that we will run for each tool Test 1 - photo-realism (a picture of mice being judged) Test 2 - following instructions (a treasure island map outline) Test 3 - being creative (a toy sheep as superhero) Summing up: how our tools compare for cost, speed, ease of use and quality Conclusions and recommendations

Comparing the 5 leading AI tools for image generaton from text

Part two of an eight-part series of blogs

Having recently compared the 4 main AI tools for text prompts, I thought I'd do the same for image generation tools. In this blog series we compare Dall-E (via ChatGPT and separately via Copilot), Firefly, Midjourney and Stable Diffusion for 3 pre-defined tests to see which ones score most highly for cost, ease-of-use, speed, editing ability and above all for quality of image.

A deep-dive review of text-to-image AI tools
The 5 AI image generation tools that we will compare (this blog)
The 3 pre-defined tests that we will run for each tool
Test 1 - photo-realism (a picture of mice being judged)
Test 2 - following instructions (a treasure island map outline)
Test 3 - being creative (a toy sheep as superhero)
Summing up: how our tools compare for cost, speed, ease of use and quality
Conclusions and recommendations

Posted by Andy Brown on 22 July 2024

In this blog

Claude
Imagen from Gemini (Google)
Canva
Generative AI from Getty Images

When choosing AI text-to-image tools, there are 4 obvious names to include:

Tool	Reason to include it
Dall-E	This OpenAI tool was the first one to allow proper generation of images from text prompts, and although it's now mostly accessed from within Microsoft's ChatGPT and Copilot, it's still the one to beat. Originally I was going to cover ChatGPT only, but I've included Copilot as it seems to give significantly different results.
Midjourney	Midjourney is an independent research laboratory based in San Francisco. It's been used to design a front cover for The Economist (in June 2022) as well as the image which won the digital art competiton at the Colorado State Fair in 2022. Midjourney has a reputation for producing the highest quality images of all of the AI text-to-image tools.
Stable Diffusion	This product from Stability AI allows open-source editing. The underlying model was trained on a library of some 5 billion images, some of which (it's being alleged in court) were privately owned.
Adobe Firefly	Any comparison of image generation tools has to include Adobe's offering, since they have dominated the world of images for so long.

Below are some notes on the tools which didn't make the cut.

Claude

Claude is great, but .... it doesn't do images:

Shame, really - maybe by the time you're reading this Anthropic will have added this feature to Claude.

Imagen from Gemini (Google)

This would have been an obvious candidate for inclusion in this review, were it not for this message on Google's website:

I'm sure this will come back bigger and better, but for now ...

Canva

I did try with this high-flying Australian company's image generator tool, but it just didn't quite seem up to scratch:

Things like this didn't inspire confidence, but my main reason for not including Canva in this review was its low profile on the Internet compared to the other tools.

Generative AI from Getty Images

The great selling-point of this tool is that they remove any risk of copyright infringement, since the tool was trained on the vast Getty image library:

The Getty strapline - but I have to confess I don't worry much about my use of the other text-to-image tools. Maybe I should?

Here's why I decided not to review this tool:

This is a different ballpark to the pricing for the other tools I've included in this review, and I can't see many readers being willing to pay these prices.

So having established what tools we're going to look at, let's now see what we'll ask each to do!

Parts of this blog
A deep-dive review of text-to-image AI tools The 5 AI image generation tools that we will compare (this blog) The 3 pre-defined tests that we will run for each tool Test 1 - photo-realism (a picture of mice being judged) Test 2 - following instructions (a treasure island map outline) Test 3 - being creative (a toy sheep as superhero) Summing up: how our tools compare for cost, speed, ease of use and quality Conclusions and recommendations

Parts of this blog

A deep-dive review of text-to-image AI tools
The 5 AI image generation tools that we will compare (this blog)
The 3 pre-defined tests that we will run for each tool
Test 1 - photo-realism (a picture of mice being judged)
Test 2 - following instructions (a treasure island map outline)
Test 3 - being creative (a toy sheep as superhero)
Summing up: how our tools compare for cost, speed, ease of use and quality
Conclusions and recommendations

Some other pages relevant to the above blogs include:

Classes in AI (Artificial Intelligence)

This blog has 0 threads

Add a new post

BLOGS BY TOPIC

BLOGS BY AUTHOR

Claude

Imagen from Gemini (Google)

Canva

Generative AI from Getty Images

Head office

London

Manchester