New integrated mode for GPT-4 in ChatGPT+
Nov. 9th, 2023 03:20 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png)
They switched me to the new integrated mode. This is supposed to have all kinds of upgrades, in particular it is possible to read and write images in one session.
I still don't know if it is possible to competently edit images via this workflow (I'll try, and I'll read what other people say, but it might be difficult without a more image-oriented input system). Nevertheless, just like in dmm.dreamwidth.org/76698.html I read my avatar icon I am using here and talk to the new GPT-4 about it, and then I ask it to produce images based on that, see the comments for conversation and images.
I still don't know if it is possible to competently edit images via this workflow (I'll try, and I'll read what other people say, but it might be difficult without a more image-oriented input system). Nevertheless, just like in dmm.dreamwidth.org/76698.html I read my avatar icon I am using here and talk to the new GPT-4 about it, and then I ask it to produce images based on that, see the comments for conversation and images.
no subject
Date: 2023-11-09 08:31 pm (UTC)(That's actually quite good.)
ME: Great, this is an awesome image! Now, can you try to remix it with the image I uploaded?
no subject
Date: 2023-11-09 08:32 pm (UTC)no subject
Date: 2023-11-09 08:34 pm (UTC)(Mmmmm... may be... the question whether one can have good control over image creation process remains open.)
no subject
Date: 2023-11-10 03:15 pm (UTC)Great! What were the prompts you have sent to DALL-E for these two images?
GPT-4:
For the first image, the prompt sent to DALL-E was:
"Visualize a futuristic neural machine, where single neurons are depicted as intricate, metallic structures with branching, tree-like cables or conduits representing flexible tree-shaped tensors. These neurons are interconnected with glowing pathways of data flow, highlighting the transformation and processing power within a complex network. The entire scene should have a luminous, high-tech feel, with a clear sense of advanced computation and artificial intelligence at work. The background should be dark to contrast the bright, energy-filled connections of the neural network."
For the second image, the prompt was:
"Combine the concept of a neural machine with the schematic design from the user's uploaded image. The image should depict a fusion of the intricate, metallic tree-like structures representing neurons with the optical lens-like elements from the schematic. The neurons should have branching, flexible tensors interwoven with the schematic's lens and ray design, creating a harmonious blend of organic and structured elements. The entire composition should be illuminated with glowing pathways of data flow, set against a dark background to emphasize the complexity and elegance of this neural optical machine."
These prompts were designed to create a visual representation that merges the abstract concept of a neural network with the schematic structure you provided.