4o ImageGen: Share your best pictures

And now: A memetic image prompted with text - and shut down by the vision inspection mid-blur during generation, for the f-word appearing in what was to be re-created in joyous quality…So a new session was required with a substitute phrasing (for a minute of typing those words in Photoshop).

Clear error-free wording without any jumble.

Follow-up corrective generations seem to lose quality in the depictions, compared to starting a new session for one image. Perhaps the image going back in for vision understanding is a degrading experience for the AI.

The internet original

6 Likes

6 Likes


4 Likes

With new ImageGen in ChatGPT-4o

The Last Spring

5 Likes

How can I go back to the old version of the shape?

2 Likes

yes, that’s what I’m trying to figure out but idk image style. i need list of image style.

@Daller You must see this:

“black panther” and “snow white” are working, there is no Disney’s Snow White or Marvel’s Black Panther. Well done! OpenAI has heard your voice! :grinning_face:

5 Likes

We can use old version still on DALL-E (now legacy), but IDK how long can we use it:

ChatGPT - DALL·E

4 Likes

As @polepole pointed out, you should use the Dall-E GPT, which is still available from OpenAI via the ChatGPT GPT Store.

@souly.et.music Since the capability was released just yesterday, there won’t be too many lists around just yet. I suggest testing the tool as we all are doing right now.

3 Likes

oh thank u for your helping​:folded_hands::hushed_face::smiling_face_with_three_hearts::blue_heart:

1 Like

Prompt:
Ultra-realistic photo of a futuristic astronaut standing on a rocky alien terrain, with a colorful, vibrant outer space background filled with nebulae, swirling galaxies, and bright stars in shades of purple, blue, pink, and gold, high detail, cinematic lighting.

5 Likes

4 Likes

Thanks to everyone at OpenAI for all your hard work.

I’ve been working on a series of images for young children, and of course saving the prompts as I go along. The differences between DALL-E 3 and the new 4o imaging is quite pronounced. For these particular images, I love the DALL-E 3 style I have been able to evoke, but dislike how the images are often impervious to my prompts as regards things to include, things to not include, misc. item details, and hallucinations. [ thank providence for Photoshop’s new Generative Fill tool for making corrections very very quickly ]

Here’s one such prompt, and the [cleaned up a bit with Photoshop] DALL-E 3 result side by side with the new engine’s output. The prompt is quite long and verbose, due to my trying to work around misc. DALL-E 3 problem areas.


A vintage-style portrait mode illustration with aspect mode 1024x1792. The top half of the image shows a snow-filled stormy sky. The top of the bottom half shows a steep mountain hill, surrounded by forest, with Mt. Shasta in the background. Snow is swirling and blowing. The bottom of the bottom half shows exactly two adorable, laughing baby emperor penguins dressed in colorful winter clothing, sledding down a steep snowy hill amidst a heavy swirling snowstorm, grabbing one another for dear life. Each penguin is adorned in vibrant vintage winter coats, knit caps, woolen scarves, warm pants, thick knit socks, and boots. They are lying prone on classic vintage Flexible Flyer wooden sleds with distinctive red runners and a large, sturdy wooden cross-member at the front for steering. The cross-member is thick, flat, and matches the width and thickness of the sled’s wooden slats. It extends about 6 inches beyond the sled’s rails on each side, designed for steering by shifting it left or right with hands or feet. The cross-member is clearly detailed, emphasizing its solid wood texture and functional design. The hill is encircled by a snow-covered forest of towering pine trees, dusted with fresh snow. In the background, the majestic higher reaches of Mt. Shasta rise, partially veiled by snow clouds. It’s early morning with a golden glow filtering through patches of blue sky. Heavy snow is falling, creating a magical winter atmosphere. The illustration features delicate ink outlines, a muted color palette, and a nostalgic aesthetic reminiscent of early 20th-century detailed illustrations for children. Created using: enhanced hand-drawn textures, fine ink detailing, watercolor, tempera, colored oil pencils, soft shading, nostalgic color grading, vintage children’s illustration aesthetics, detailed fabric textures, early 20th-century print design, warm and gentle lighting, high-definition quality, natural aesthetic, with no framing elements, words, letters, or banners.


I love the new image, but it is clearly from a very different universe. My one disappointment: the inability to create an image in the same 9:16 aspect ratio I’ve been using with DALL E 3. Hopefully that capability will manifest Real Soon Now.

Clearly there is much new stuff to explore and learn. I’ll keep using DALL-E 3 for this particular project, but will start most new projects using the new stuff. Love my magical AI image tools ! [ says the artist of 55+ years practice, during which I’ve used all sorts of image tools ] Endless learning and joy !

6 Likes

Two things on my wish list:

  1. The ability to keep using DALL E 3 for a very long time. I’ve got a major investment in prompts that come out a particular way with DALL E 3, and hate to lose their utility. [ One of the things I liked best while working at MS in the Systems div. back in the 80s and 90s: the company’s obsession with keeping older software running on new versions of the OSes. A win.win.win for developers, end-users, and MS. ]

  2. The ability to feed ChatGPT an image produced using DALL E 3, and have it spit out a prompt that will create as closely as possible that same image using the new 4o model, esp. as regards styling.

1 Like

It took me very little time to be annoyed with the fake image loading fade-in effect, even switching between chats.

You are simply sent a few mid-generation images that could be displayed as soon as they are received.

Here is a Tampermonkey script with CSS-as-Javascript that will knock it off. I hit just about everything that attempts such fading or blurring or transitions, with little testing elsewhere in ChatGPT, so try at your peril (peril of losing other laggy fades also), with a “disable script” button always available.

// ==UserScript==();

(OpenAI busted this within two days...)
3 Likes

original:

Recreated version:

2 Likes

Negation prompts work well:

5 Likes

I’d love to show mine but image generation has been down since the new model last night - is there some trick to it? Chat just says image generation is down.

8 Likes