Multimodal AI processes different types of information, such as text, images or audio. Google gives examples of models receiving an image and producing a written response. The term describes the types of information involved, rather than specifying every format that a particular product supports. Source: Google Cloud: Multimodal AI

Ask for a relevant demonstration

Choose a sample document or image that reflects the intended task and contains no sensitive traveller information. Ask the supplier to show the supported input and expected output. This makes the demonstration useful for the workflow being considered.

Check the result against the original

If the task extracts information from an itinerary image, compare names, dates and conditions with the original. Record unclear or missing material for a person to resolve. This gives the team a practical acceptance check before using the extracted information elsewhere.

Written and updated by Travelgenix. Sources checked on 19 October 2026. Practical advice and examples are Travelgenix editorial guidance.