What decides if a model can answer questions about a photo

Gen AI Leader Fundamentals of gen AI Easy Question 4 of 60

A field-inspection app must accept a photograph of a damaged part and answer a technician's typed question about what it shows.

Which model characteristic decides whether a model can serve this?

Show answer and explanation

DThe input modalities the model accepts

Modality is the property that says which input types a model can take - text only, or text plus images, audio or video. A model that cannot accept an image cannot serve this app at any context length. The context window governs how much information fits in one request, which matters once the model can read the photo at all, not before. The knowledge cutoff describes what the model absorbed in training and is irrelevant to a photograph supplied in the request. The output limit caps how long the answer runs. Whichever model is chosen, check the input types supported by the specific deployment rather than assuming the family supports them.

Source: services.google.com/fh/files/misc/generative_ai_leader_exam…

Anonymous, no account needed. Your tally stays in this browser.

Next question Start 5 timed questions Browse all 60 original Gen AI Leader practice questions