Yep, the point we wanted to make here is that GPT-5.2's vision is better, not perfect. Cherrypicking a perfect output would actually mislead readers, and that wasn't our intent.

And here is Gemini 3: https://media.licdn.com/dms/image/v2/D5610AQH7v9MtrZxxug/ima...

Promotional content for LLMs is really poor. I was looking at Claude Code and the example on their homepage implements a feature, ignoring a warning about a security issue, commits locally, does not open a PR and then tries to close the GitHub issue. Whatever code it wrote they clearly didn't use as the issue from the prompt is still open. Bizarre examples.

Also a "stacked pair" of USB type-A ports, when there are clearly 4

Not that bad compared to product images seen on AliExpress.

General purpose LLMs aren't very good with generating bounding boxes, so with that context, this is actually seen as decent performance for certain use cases.

FTA: Both models make clear mistakes, but GPT‑5.2 shows better comprehension of the image.

You can find it right next to the image you are talking about.

to be fair that image has the resolution of a flip phone from 2003

Because the whole culture of AI enthusiasts is to just generate slop and never check the results

You seen the charts on their last release? They obviously don’t check - too rich