# What Is TTFT (time to first token)?

> TTFT (time to first token) is the time from sending a request to receiving the model's first token, and it decides how responsive a service feels. It is dominated by the prefill phase: the model must process the entire input before it can start producing output.

URL: https://vuke.top/en/glossary/ttft · Updated November 12, 2026

TTFT (time to first token) is the time from sending a request to receiving the model's first token, and it decides how responsive a service feels. It is dominated by the prefill phase: the model must process the entire input before it can start producing output.

## In the measurements

Once the KV cache of a long-context session is evicted, TTFT grows sharply. In measurements, TTFT p50 is 149.5 s with full recompute and 11.85 s when reading back from FX100 (480B model, concurrency 16) [R2](https://vuke.top/en/evidence/R2).

Source: [R2](https://vuke.top/en/evidence/R2)