DALL-E 3: Release Date, Features and Comparison
After the release of DALL.E in January 2021 and the eventual release of its second version in 2022, the tool has bagged over 1.5 users actively creating 2 million images per day— DALLE.3 is the latest version of OpenAI’s AI image synthesis model which features complete integration with ChatGPT.
The earlier version of DALL.E— being one of the first text-to-image generative AI tools, faced mixed reviews from users who had complained of the images being “less detailed” and “distorted”.
It procured photorealistic explicit images and provided AI bias in its results, but DALL.E 3 understands significantly more nuance and detail than the previous version. You can now easily translate ideas into exceptionally accurate images.
You can directly create images from ChatGPT and obtain accurate representation of any scene or object but here’s what has improved.
What’s New with DALL.E 3?
- Accurate human details for eg: hands and eyes
- No hacks or prompt engineering is required
- Available to ChatGPT Plus and Enterprise customers
- Focus to Detail
- More realistic
Even with the same prompt, DALL·E 3 delivers significant improvements over DALL·E 2.
- Prompt: “An expressive oil painting of a basketball player dunking, depicted as an explosion of a nebula.”
DALL-E 3 can interpret the content within context better than its previous versions and comes with more safety insurance.
The open-source API is currently in public beta and available for testing (limited by number).
According to OpenAI, the team has put effort into building strong safety features to head off elicited and bigoted imagery.
The company used external red teams and leaned into input classifier techniques for training a language model to disregard certain words to prevent explicit or violent prompts.
OpenAI took to X on 20th September and announced the new release
“Our new text-to-image model, DALL·E 3, can translate nuanced requests into extremely detailed and accurate images. Coming soon to ChatGPT Plus & Enterprise, which can help you craft amazing prompts to bring your ideas to life” – OpenAI, 2023
In order to help keep the lawyers away from its doorstep, OpenAI is also offering artists an opportunity to have their work excluded from subsequent iterations of the company’s text-to-image AI models.
Creators can submit an image that they have the right of ownership over and file the removal on Form (on its website).
DALL-E 3 has taken full responsibility for filtering out images that look similar to what the artist drew and how they drew them.
This case was considered because OpenAI doesn’t want to raise yet another question of who owns this art.
Artists in the past have filed lawsuits against DALL-E rivals Stability AI and Midjourney, as well as art website DeviantArt, in response to the training of their copyrighted images on text-to-image models.
According to OpenAI’s official website, “DALL·E 3 is designed to decline requests that ask for an image in the style of a living artist. Creators can now also opt their images out from training of our future image generation models.”
DALLE-3 Release Date
DALL-E 3, an AI image generator, was initially launched on August 20, 2023, exclusively for researchers. In October, the new version of DALL-E 3 saw a wider release and is now available publicly but only through Microsoft’s Bing Chat. Additionally, this feature is accessible exclusively to ChatGPT Plus users.
How to Use DALL-E 3?
1. Sign up for ChatGPT Plus
Currently, DALL·E 3 is exclusive to ChatGPT Plus users. To use it, first create a ChatGPT account. Then, go to the bottom of the left sidebar and click on ‘Upgrade to Plus’. Choose the ChatGPT Plus plan for $20/month, add your payment info, and you’re set.
You can experiment with DALL·E 3 in Bing Image Creator without needing ChatGPT Plus. It’s based on the same model. The experience is a bit different; less control and images come with watermarks.
It operates on a “boosts” credit system, which is technically free. Once your credits are used up, expect slower render times for your prompts in Image Creator.
2. Write Text Prompt
Using DALL·E 3 is similar to ChatGPT. Type a text prompt or chat naturally. Each prompt generates four unique images to pick from.
The greatest thing about DALL-E 3 is that you don’t have any word limit to how long you want a text prompt
- “A watercolor illustration of a cat playing a violin in a city alley at night.”
- “A digital art depiction of a futuristic cityscape with flying cars and neon lights.”
- “A pencil sketch of a woman reading a book under a large oak tree in a serene park.”
- “A surrealistic painting of a clock melting over an ancient Greek temple.”
- “A chalk pastel drawing of a Viking ship sailing under the Northern Lights.”
3. Get Unique AI Images
You will be offered 1 AI image that you can download; hover over it, and click the download icon in the top-left.
DALLE.3 Vs. Midjourney
1. The Human Heart Test
Prompt: “An illustration of a human heart made of translucent glass, standing on a pedestal amidst a stormy sea. Rays of sunlight pierce the clouds, illuminating the heart, revealing a tiny universe within. The quote ‘Find the universe within you’ is etched in bold letters across the horizon.”
2. Architect and Buildings
Prompt: “A modern architectural building with large glass windows, situated on a cliff overlooking a serene ocean at sunset.”
3. Imagination and Creative Details
Prompt: “Tiny potato kings wearing majestic crowns, sitting on thrones, overseeing their vast potato kingdom filled with potato subjects and potato castles.”
4. The Ceramic Maiden
Prompt: “A middle-aged woman of Asian descent, her dark hair streaked with silver, appears fractured and splintered, intricately embedded within a sea of broken porcelain. The porcelain glistens with splatter paint patterns in a harmonious blend of glossy and matte blues, greens, oranges, and reds, capturing her dance in a surreal juxtaposition of movement and stillness. Her skin tone, a light hue like the porcelain, adds an almost mystical quality to her form.”
5. Melodious Leaf Ensemble
Prompt: “A 2D animation of a folk music band composed of anthropomorphic autumn leaves, each playing traditional bluegrass instruments, amidst a rustic forest setting dappled with the soft light of a harvest moon.”
6. Innovative Chairs
Prompt: “Photo of a lychee-inspired spherical chair, with a bumpy white exterior and plush interior, set against a tropical wallpaper.”
7. Soulful Dancer
Prompt: “In front of a deep black backdrop, a figure of middle years, her Tongan skin rich and glowing, is captured mid-twirl, her curly hair flowing like a storm behind her. Her attire resembles a whirlwind of marble and porcelain fragments. Illuminated by the gleam of scattered porcelain shards, creating a dreamlike atmosphere, the dancer manages to appear fragmented, yet maintains a harmonious and fluid form.”
Final Thoughts
We had a hard time choosing between the two images, as they both matched the prompt well.
However, we decided to do some research on the hermit crab and found out that DALL-E 3’s image was more accurate in terms of the crab’s anatomy.
The DALL-E 3 image looked more like a real hermit crab than the Midjourney image.
DALL.E 3 like its previous versions will not be free and may cost credits to craft your imagination.
It will be available with ChatGPT, Bing, and Microsoft Designer.
Are you excited about the upcoming version of DALL.E? Do you think it will be able to actively compete with MidJourney?
