LLM Reference
TTFAactiveAudio

TTFA: Time to First Audio (ms)

Metric: ms (lower is better)Introduced: 2024

Time in milliseconds from the end of a user utterance until the first audio byte is received from the model, measured at representative API endpoints by Artificial Analysis (P50). Sub-300 ms is generally imperceptible; 200 ms matches natural human turn-taking. Lower is better.

Models ranked

4

tracked on this benchmark

Score band

630.0 – 1140.0

best → lowest tracked

Snapshot trend

need ≥2 snapshots

Leaderboard

Tracked models ranked by ms (lower is better).

Compare candidates
#Model variant and provenanceScore
1
Gemini 2.5 Flash Live API
Version: p50-msHarness: Not recordedEvaluator: Not recordedObserved: Jun 7, 2026Confidence: Not recordedSource
630.0
2
Grok Voice Think Fast 1.0
Version: p50-msHarness: Not recordedEvaluator: Not recordedObserved: Jun 7, 2026Confidence: Not recordedSource
780.0
3
gpt-realtime-1.5
Version: p50-msHarness: Not recordedEvaluator: Not recordedObserved: Jun 7, 2026Confidence: Not recordedSource
820.0
4
Amazon Nova 2 Sonic
Version: p50-msHarness: Not recordedEvaluator: Not recordedObserved: Jun 7, 2026Confidence: Not recordedSource
1140.0

How to read this benchmark

This benchmark scores models where lower is better. Use scores for directional filtering and shortlisting, not universal quality ranking; then validate pricing, context window, provider availability, and fit for your workload.

Trust this score when

  • There is a fresh timestamped snapshot (or multiple snapshots) for this benchmark.
  • The model list covers the same version family you can actually deploy today.
  • Top candidates overlap with your required routing and feature requirements.

Be cautious when

  • There is only one benchmark snapshot or the dataset appears stale.
  • The benchmark metric direction is opposite of your decision objective.
  • The score difference between options is narrow and likely within implementation variance.

Related benchmarks

Last reviewed: Jun 7, 2026

Resources