Better view for generated images

[nepalevov]

Description

I would like to see generated images from models with image-generation capability immediately, without additional actions.

Current state

As a response, we receive a "Image" attachment-like spoiler frame, closed by default.
The user have to open the spoiler manually
NOW the image loading starts
The image looks huge and loads at full resolution; no optimizations; it takes time and depends on network load. The average raw image size is ~2Mb

Use case/motivation

Currently, I have to open a spoiler manually each time, which is very inconvenient. Especially, when the ONLY expected output from the model is a picture, e.g. "Stable Diffusion" or "DALL-E".

I expect one of the denoted below (or similar):

1. best option: a mini-gallery of picture(s) as a response content. Relatively small grid(?) of pictures, maybe thumbnails with a full view on a click
2. compromise option: the picture is embedded into response content without an "Image" attachment-like spoiler frame
3. minimum effort option: everything as is, but the "Image" attachment-like spoiler frame is opened by default

Things to think about:

1. progressive image loading instead of default top-to-bottom
2. thumbnails with decreased quality
3. ???

Examples

OpenAI

Google

Microsoft

Related issues

No response

Confidential information

• I confirm that do not share any confidential information