jargon

Applied AI·The API layer

the long prompt made the wait before the first word noticeably worse, even though generation ran at the same speed after that.

Time to first token (TTFT)

Also calledTTFT

Draft summary, pending review

How long before the first output token arrives. Dominated by input processing, so long prompts raise it. The metric that matters most for perceived responsiveness.

Commonly confused with