728 ms median and 789 ms P95 time to first audio on Coval (30 days to October 22, 2026): Visible delay speed tier.
Benchmark summary: Visible delay speed tier on Coval's P95 (the voice becomes a visible part of the delay).
Alibaba Qwen3 TTS Flash Realtime is a proprietary text-to-speech (TTS) API from Alibaba (Qwen), measured by one independent benchmark, Coval.
Speed tier
Visible delayP95 of 320 ms or more. The voice becomes a visible part of the delay.
Data read October 22, 2026. No combined score: methodology
Gradium TTS vs Alibaba Qwen3 TTS Flash Realtime Fastest TTS APIs, by speed tierOn Coval's 30-day window ending October 22, 2026 (3,366 runs from US East), Alibaba Qwen3 TTS Flash Realtime starts speaking after 727.7 ms on a typical turn (median) and 789.2 ms on a slow turn (P95, the slowest one in twenty). That P95 is 28th lowest of the 32 models Coval measures, which puts it in the Visible delay tier: the voice becomes a visible part of the delay.
Neither Speko nor Artificial Analysis rates its voice quality.
Speko has not scored its pronunciation of numbers and dates.
Coval lists $13 per 1M characters (since Sep 1, 2026). List prices, as each source prints them.
Smallest.ai Lightning V3.1 Pro (Jul 2026) is a proprietary text-to-speech (TTS) API from Smallest.ai, measured by three independent benchmarks: Artificial Analysis, Coval and Speko.
Google Chirp 3: HD is a proprietary text-to-speech (TTS) API from Google, measured by two independent benchmarks: Artificial Analysis and Coval.
Murf AI Falcon 2 is a proprietary text-to-speech (TTS) API from Murf AI, measured by two independent benchmarks: Artificial Analysis and Coval.