click any column header to sort

TTS Bench — Speed — windows-cloning

Rig: windows-5090 — AMD Ryzen 9 9950X3D 16-Core Processor (16C) · NVIDIA GeForce RTX 5090 32GB · 126 GB RAM · Windows 11
Label: 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): StyleTTS 2 (cuda) — 41.66× warm RTFx, 212ms warm TTFA

ModelDeviceTTFA coldTTFA warmRTFx coldRTFx warmPeak RAMPeak VRAMSizeReleased
StyleTTS 2cuda1.66s212ms5.45×41.66×2.63 GB1.52 GB~148MJun 2023
StyleTTS 2cpu1.70s218ms5.27×40.62×2.63 GB~148MJun 2023
OpenVoice v2cuda1.72s361ms3.61×16.73×2.65 GB1.35 GB~100MApr 2024
Sopro V2 Turbocuda738ms558ms10.31×14.41×2.46 GB993 MB120MAug 2026
Audio8 TTS 0.6B (compiled)cuda1.13s610ms6.09×11.69×2.48 GB2.71 GB601MAug 2026
OmniVoicecuda1.20s803ms6.67×9.75×2.27 GB2.42 GB~1BMar 2026
Sopro V2 Turbo (streaming)cuda228ms120ms7.77×9.55×2.48 GB1.22 GB120MAug 2026
MiraTTScuda2.14s1.46s4.76×7.01×3.58 GB1.28 GB0.5BDec 2025
LongCat-AudioDiT 1Bcuda1.66s1.36s4.72×5.67×2.36 GB5.78 GB1.42BMar 2026
F5-TTS v1cuda1.56s1.06s3.75×5.50×2.67 GB883 MB330MOct 2024
Coqui XTTS-v2cuda2.02s1.51s3.02×4.76×2.35 GB2.10 GB750MOct 2023
OpenVoice v2cpu2.60s1.44s2.23×4.25×2.78 GB~100MApr 2024
Chatterbox Turbocuda2.66s1.59s2.43×4.15×2.69 GB3.33 GB744MDec 2025
Pocket-TTScpu136ms124ms4.11×4.10×2.29 GB100MJan 2026
Qwen3-TTS 1.7B (CUDA-graph)cuda7.93s1.57s0.83×3.66×2.61 GB4.89 GB1.7BJan 2026
LongCat-AudioDiT 3.5Bcuda2.59s2.32s3.13×3.49×2.26 GB15.14 GB3.83BMar 2026
Echo-TTScuda2.84s2.23s2.62×3.39×1.94 GB9.38 GB2.8BDec 2025
Chatterboxcuda4.01s2.88s1.63×2.33×3.02 GB3.63 GB1.2BApr 2025
DramaBoxcuda6.46s4.99s1.66×2.14×2.45 GB17.39 GB3.3BApr 2026
WavTTS 0.67Bcuda3.16s2.84s1.85×2.08×2.00 GB1.44 GB0.67BMay 2026
NeuTTS Nanocuda854ms331ms1.60×2.00×2.59 GB3.26 GB229MDec 2025
Sopro V2 Turbocpu4.45s4.46s1.76×1.73×1.41 GB120MAug 2026
Sopro V2 Turbo (streaming)cpu888ms816ms1.67×1.69×1.59 GB120MAug 2026
MOSS-TTS-Nanocuda5.59s4.87s1.34×1.66×2.42 GB910 MB100MApr 2026
VibeVoice 7Bcuda5.22s4.86s1.41×1.58×2.04 GB17.63 GB7BSep 2025
NeuTTS Nanocpu863ms387ms1.23×1.58×5.12 GB229MDec 2025
IndexTTS-2cpu5.77s4.87s1.20×1.47×5.54 GB1.5BJun 2025
MOSS-TTS v1.0cuda5.44s4.73s1.31×1.44×2.10 GB22.83 GB8BFeb 2026
VoxCPM2 2Bcuda6.36s4.72s0.96×1.33×2.49 GB5.71 GB2BApr 2026
NeuTTS Aircuda1.46s492ms0.88×1.30×2.96 GB3.26 GB748MSep 2025
IndexTTS-2cuda6.80s5.66s1.03×1.26×4.05 GB7.71 GB1.5BJun 2025
MOSS-TTS-Nanocpu5.69s5.04s1.16×1.25×3.05 GB100MApr 2026
MOSS-TTS v1.5cuda5.88s5.28s1.14×1.24×2.36 GB22.83 GB8BMay 2026
NeuTTS Aircpu1.75s568ms0.75×0.96×5.47 GB748MSep 2025
Fish Speech 1.5cuda8.04s8.21s0.85×0.95×3.82 GB1.76 GB~500MNov 2024
Coqui XTTS-v2cpu9.93s8.76s0.84×0.85×3.43 GB750MOct 2023
Zonos v0.1cuda10.36s9.85s0.67×0.73×6.55 GB4.45 GB1.6BFeb 2025
Audio8 TTS Preview 0.6Bcuda13.33s10.65s0.58×0.73×3.01 GB2.75 GB601MJul 2026
Chatterbox Turbocpu10.84s9.37s0.68×0.71×4.39 GB744MDec 2025
ZipVoice 123M (4/5 ok)cuda68.19s105.99s0.38×0.60×25.56 GB53.24 GB123MJun 2025
Sesame CSM-1Bcuda13.21s11.46s0.47×0.55×2.47 GB3.54 GB1BMar 2025
Dia 1.6B-0626cuda14.04s11.22s0.33×0.47×8.53 GB4.82 GB1.6BJun 2025
Breeze TTS 2cuda1.22s400ms0.43×0.45×2.45 GB7.62 GB3.47BAug 2026
Audio8 TTS Preview 0.1Bcuda20.61s17.58s0.35×0.45×2.93 GB1.86 GB170MAug 2026
ZipVoice 123Mcpu50.03s47.06s0.30×0.43×39.66 GB123MJun 2025
VoxCPM2 2Bcpu18.12s16.04s0.34×0.39×7.53 GB2BApr 2026
Chatterboxcpu17.39s17.39s0.36×0.38×4.72 GB1.2BApr 2025
Miso TTS 8Bcuda37.67s32.57s0.30×0.32×2.07 GB16.55 GB8.2BMay 2026
OuteTTS 1.0 1Bcuda18.32s17.44s0.30×0.30×3.64 GB3.56 GB1BApr 2025
Audio8 TTS Preview 0.1Bcpu38.30s37.62s0.18×0.19×4.59 GB170MAug 2026
Fish Speech 1.5cpu37.45s42.66s0.18×0.18×4.43 GB~500MNov 2024
Audio8 TTS Preview 0.6Bcpu47.09s45.60s0.15×0.16×6.64 GB601MJul 2026
OmniVoicecpu54.74s53.28s0.14×0.14×3.98 GB~1BMar 2026
Zonos v0.1cpu61.60s60.84s0.12×0.12×8.57 GB1.6BFeb 2025
Sesame CSM-1Bcpu48.90s48.64s0.12×0.12×6.32 GB1BMar 2025
Mars5-TTScpu41.89s43.23s0.08×0.10×2.24 GB1.2BJun 2024
F5-TTS v1cpu71.35s71.18s0.08×0.08×2.71 GB330MOct 2024
Mars5-TTScuda41.12s39.53s0.09×0.07×2.23 GB7.23 GB1.2BJun 2024
LuxTTScpuLuxTTS install failed (piper-phonemize has no Windows wheels)