File talk:Artificial intelligence DALL-E generated images from the prompt Listenbourg.png

来自Wikimedia Commons
跳转到导航 跳转到搜索

The source of the image specified in The New Zealand Herald is "Photo / Dall E, Creative commons" (archive).

Maybe it's DALL-E Mini. Read DALL-E#Open-source implementations: "Released in 2022 on Hugging Face's Spaces platform, Craiyon (formerly DALL-E Mini until a name change was requested by OpenAI in June 2022) is an AI model based on the original DALL-E that was trained on unfiltered data from the Internet."

Many newspapers like CNBC] mix DALL-E and DALL-E Mini.

In doubt, the Category:Text-to-image generation is correct. -- Tukp (留言) 08:42, 12 March 2023 (UTC)[回复]

I am pretty sure that The New Zealand Herald was referring to DALL-E Mini when they said "DALL-E", the layout is very clearly from DALL-E Mini. I think it's pretty safe to put it in the Craiyon category. IsraeliEditor54 (留言) 08:43, 12 March 2023 (UTC)[回复]
Please read COM:NPOV and en:WP:STICKTOSOURCE. -- Tukp (留言) 09:01, 12 March 2023 (UTC)[回复]
This is also evident by the 9 images generated, when DALL-E only does 4. Also, the orange accent color makes it somewhat obvious. Wiichicken (留言) 01:23, 30 June 2024 (UTC)[回复]
Ok so I guess it's safe to leave it in Category:Text-to-image generation, however someone just removed it from that category. IsraeliEditor54 (留言) 12:30, 13 March 2023 (UTC)[回复]