Multimodal Model
Generative AI & LLMsChapter 4: Choosing the LLMTerm 84 of 253
Model that understands/generates multiple data types: text, images, audio, video.
At Sunrise School
Gopal asksA parent sends a photo of a torn report card.
Bit answersMultimodal models read images and text together. I see the photo, read the student's name, and explain how to get a duplicate.
In the real worldUpload a screenshot of an ST22 dump and ask why.
Watch the short
Short coming soonSeason 2 · Day 20Follow @haigopala
Short coming soonSeason 2 · Day 20Follow @haigopala.telugu
Short coming soonSeason 2 · Day 20Follow @haigopala.hindi
Quick check
Quick check. Can a text-only model read a photo? Yes or no?
No! Reading images needs a multimodal model.
Practise it
Try this term and all 253 others in the free Model Training Arena.
Open the Arena →Follow Bit daily: YouTube English · తెలుగు · हिंदी | Instagram English · తెలుగు · हिंदी