"only AI Model with Vision-to-Language Capabilities" What do they mean by this? ... | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		jl6 on May 13, 2024 \| parent \| context \| favorite \| on: Falcon 2 "only AI Model with Vision-to-Language Capabilities" What do they mean by this? Isn't this roughly what GPT-4 Vision and LLaVA do?

Hugsun on May 13, 2024 | [–]

At first I thought they were playing some semantic game.

Something like LLaVA being a language to vision model but I can't steelman the idea so it makes sense.

Maybe they're just lying?

tictacttoe on May 13, 2024 | [–]

And all Claude models…

Me1000 on May 13, 2024 | [–]

And Gemini.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact