In agentic use cases, we save on end-to-end latency by spending more time and compute on individual searches. This happens because agents do fewer searches, use fewer tokens, and end up using fewer thinking tokens when using the Parallel Search API.
How would the product work if for example I just wanted to get basic information from various websites, such as "how many calories are in a McDonald's hamburger?" I'm creating a product that tracks calories via natural language and I'd want a structured JSON back of calories, grams of fat, protein, carbs etc. What API would I use for that, the Chat API one?