Gemini 3.1 Flash TTS Preview
Cost-efficient, expressive, and steerable text-to-speech model.
Google's preview text-to-speech model for expressive audio generation, with granular audio tags, support for 70+ languages, and SynthID watermarking.
Pricing
Artificial Analysis · Apr 25, 2026, 1:00 PM
Input / 1M
$2.00
Output / 1M
$12.00
Blended / 1M
$4.50
Output TPS
0
TTFT (s)
0
Model Intelligence
Arena ranking
31
Benchmarkable
Yes
Model level
release
Intelligence Index
31.1
Coding Index
24.6
Math Index
78.3
MMLU Pro
0.84
GPQA
0.79
HLE
0.13
LiveCodeBench
0.71
SciCode
0.41
AIME 2025
0.78
IFBench
0.52
LCR
0.64
TerminalBench Hard
0.17
TAU2
0.46
Recent stories
1 linked story