click any column header to sort

TTS Bench — Speed — windows-es-cloning

Rig: windows-5090 — AMD Ryzen 9 9950X3D 16-Core Processor (16C) · NVIDIA GeForce RTX 5090 32GB · 126 GB RAM · Windows 11

Label: Spanish cloning - chris_hemsworth_15s — ref chris_hemsworth_15s.wav

TTFA = time to first audio (ms; lower is better). RTFx = realtime speed (× realtime; higher is better; e.g. 10× means 10 sec of audio generated per 1 sec of compute). Cold = first run after process start; warm = subsequent runs.

Speed winners

Fastest (this cloning run): OpenVoice v2 (cuda) — 27.19× warm RTFx, 274ms warm TTFA

Model	Device	TTFA cold	TTFA warm	RTFx cold	RTFx warm	Peak RAM	Peak VRAM	Size	Released
OpenVoice v2	cuda	8.06s	274ms	0.92×	27.19×	2.62 GB	705 MB	~100M	Apr 2024
Audio8 TTS 0.6B (compiled)	cuda	1.50s	652ms	5.25×	11.19×	2.56 GB	2.35 GB	601M	Aug 2026
OpenVoice v2	cpu	2.17s	1.57s	3.43×	4.66×	1.96 GB	—	~100M	Apr 2024
Coqui XTTS-v2	cuda	17.96s	1.74s	0.44×	4.19×	2.30 GB	1.90 GB	750M	Oct 2023
Qwen3-TTS 1.7B (CUDA-graph)	cuda	22.94s	2.03s	0.28×	3.54×	2.56 GB	4.84 GB	1.7B	Jan 2026
Pocket-TTS	cpu	487ms	464ms	0.94×	1.14×	2.30 GB	—	100M	Jan 2026
Audio8 TTS Preview 0.6B	cuda	27.51s	10.39s	0.26×	0.74×	3.00 GB	2.32 GB	601M	Jul 2026
Coqui XTTS-v2	cpu	10.05s	17.21s	0.67×	0.67×	2.99 GB	—	750M	Oct 2023
Audio8 TTS Preview 0.1B	cuda	28.01s	12.02s	0.27×	0.58×	2.24 GB	1.57 GB	170M	Aug 2026
Audio8 TTS Preview 0.1B	cpu	30.94s	28.07s	0.25×	0.26×	3.91 GB	—	170M	Aug 2026
Audio8 TTS Preview 0.6B	cpu	48.16s	44.88s	0.15×	0.15×	5.33 GB	—	601M	Jul 2026
NeuTTS Nano	cuda	833ms	171ms	0.02×	0.12×	2.51 GB	3.26 GB	229M	Dec 2025
NeuTTS Nano	cpu	667ms	208ms	0.03×	0.10×	5.04 GB	—	229M	Dec 2025