AI glossary
Streaming
Streaming sends a model's reply piece by piece as it is generated, so a user sees the first words in a moment instead of waiting for the whole answer.
Most APIs stream with server-sent events: the connection stays open and each event carries the next few tokens. The page appends them as they arrive.
Streaming does not make the model faster; it makes the wait feel shorter, because reading starts while writing continues.
Stream text you show to a person. For JSON your code must parse, it is usually simpler to wait for the whole reply, or to handle partial pieces with care.
Learn it hands-on
These free lessons teach it with commands to run in a lab and checked tasks. Read them whole with no account.