TTS Bench — Speed — windows-default

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

Label: default voice

TTFA = time to first audio (ms; lower is better). RTF = real-time factor (× 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 predefined-voice: Kokoro (cuda) — 103.68× warm RTF, 67ms warm TTFA

Fastest cloning-capable: StyleTTS 2 (cpu) — 33.76× warm RTF, 251ms warm TTFA

Model	Device	TTFA cold	TTFA warm	RTF cold	RTF warm	Peak RAM	Peak VRAM	Size
Kokoro	cuda	895ms	67ms	8.32×	103.68×	2.43 GB	925 MB	82M
MeloTTS	cuda	1.61s	117ms	4.60×	60.41×	2.89 GB	1.16 GB	~52M
Piper	cpu	160ms	107ms	38.94×	58.83×	470 MB	—	~25MB
StyleTTS 2	cpu	1.74s	251ms	4.93×	33.76×	2.61 GB	—	~148M
StyleTTS 2	cuda	1.76s	265ms	4.80×	32.15×	2.60 GB	1.49 GB	~148M
OpenVoice v2	cuda	1.85s	361ms	3.65×	16.93×	2.65 GB	1.35 GB	~100M
Kokoro	cpu	609ms	532ms	12.13×	14.38×	1.82 GB	—	82M
Supertonic 3	cpu	744ms	741ms	9.86×	9.92×	570 MB	—	99M
MeloTTS	cpu	1.88s	794ms	3.68×	9.40×	2.48 GB	—	~52M
OmniVoice	cuda	1.12s	757ms	6.44×	9.30×	2.07 GB	2.16 GB	~1B
LongCat-AudioDiT 1B	cuda	1.76s	1.19s	5.10×	6.69×	2.22 GB	5.71 GB	1.42B
KittenTTS Nano 0.1	cpu	1.22s	1.20s	6.40×	6.39×	338 MB	—	<100M
F5-TTS v1	cuda	1.31s	845ms	3.47×	5.32×	2.67 GB	802 MB	330M
Coqui XTTS-v2	cuda	2.05s	1.87s	3.63×	4.75×	2.10 GB	2.14 GB	750M
Chatterbox Turbo	cuda	2.39s	1.62s	2.80×	4.28×	2.44 GB	3.01 GB	744M
LongCat-AudioDiT 3.5B	cuda	2.27s	1.80s	3.41×	4.24×	2.22 GB	14.96 GB	3.83B
OpenVoice v2	cpu	2.83s	1.48s	2.10×	4.10×	2.78 GB	—	~100M
Pocket-TTS	cpu	147ms	123ms	3.99×	4.06×	1.95 GB	—	100M
Soprano 1.1 80M	cuda	1.74s	1.77s	3.77×	3.76×	2.12 GB	326 MB	80M
Qwen3-TTS 1.7B (CUDA-graph)	cuda	6.51s	1.60s	0.90×	3.76×	2.48 GB	4.89 GB	1.7B
Echo-TTS	cuda	2.83s	2.15s	2.63×	3.44×	1.94 GB	9.38 GB	2.8B
Soprano 1.1 80M	cpu	1.97s	2.00s	3.40×	3.40×	1.34 GB	—	80M
NeuTTS Nano	cuda	678ms	258ms	2.19×	2.76×	3.24 GB	3.26 GB	229M
DramaBox	cuda	5.14s	3.78s	1.93×	2.58×	2.36 GB	17.39 GB	3.3B
VibeVoice Realtime 0.5B	cuda	3.80s	3.77s	2.24×	2.39×	1.88 GB	2.62 GB	0.5B
Chatterbox	cuda	3.35s	2.61s	1.66×	2.24×	2.80 GB	3.24 GB	1.2B
NeuTTS Nano	cpu	698ms	303ms	1.73×	2.00×	5.03 GB	—	229M
Magpie-TTS	cuda	5.48s	4.48s	1.53×	1.93×	3.54 GB	5.60 GB	357M
NeuTTS Air	cuda	1.15s	417ms	1.34×	1.62×	3.60 GB	3.26 GB	748M
MOSS-TTS-Nano	cuda	5.83s	4.92s	1.36×	1.61×	2.42 GB	777 MB	100M
VibeVoice 1.5B	cuda	4.99s	5.32s	1.44×	1.61×	2.04 GB	5.26 GB	1.5B
VibeVoice 7B	cuda	5.67s	6.17s	1.39×	1.55×	2.05 GB	17.63 GB	7B
MOSS-TTS v1.0	cuda	5.23s	4.74s	1.34×	1.48×	2.08 GB	22.83 GB	8B
VoxCPM2 2B	cuda	5.36s	5.19s	1.35×	1.35×	6.18 GB	5.65 GB	2B
IndexTTS-2	cpu	6.48s	5.42s	1.09×	1.31×	5.56 GB	—	1.5B
NeuTTS Air	cpu	1.20s	471ms	1.16×	1.29×	5.37 GB	—	748M
LFM2.5-Audio 1.5B	cpu	391ms	84ms	1.02×	1.26×	2.25 GB	—	1.5B
MOSS-TTS-Nano	cpu	6.38s	5.64s	1.08×	1.21×	3.14 GB	—	100M
LFM2.5-Audio 1.5B	cuda	410ms	91ms	0.99×	1.18×	2.25 GB	3.18 GB	1.5B
IndexTTS-2	cuda	7.20s	6.03s	0.93×	1.11×	5.88 GB	7.60 GB	1.5B
Parler-TTS Mini v1	cuda	8.29s	8.14s	0.96×	1.05×	3.19 GB	2.63 GB	878M
MOSS-TTS v1.5	cuda	7.66s	6.86s	0.88×	0.99×	2.38 GB	22.83 GB	8B
Fish Speech 1.5	cuda	8.53s	7.84s	0.81×	0.94×	3.83 GB	1.80 GB	~500M
Coqui XTTS-v2	cpu	9.92s	9.73s	0.87×	0.88×	3.23 GB	—	750M
Chatterbox Turbo	cpu	10.65s	9.92s	0.70×	0.73×	4.16 GB	—	744M
Zonos v0.1	cuda	10.88s	10.38s	0.66×	0.71×	6.51 GB	4.48 GB	1.6B
Qwen3-TTS 1.7B Base	cuda	10.77s	8.95s	0.55×	0.70×	2.42 GB	4.64 GB	1.7B
VibeVoice Realtime 0.5B	cpu	14.49s	13.35s	0.63×	0.66×	5.89 GB	—	0.5B
ZipVoice 123M (4/5 ok)	cuda	68.83s	86.51s	0.35×	0.60×	25.87 GB	53.16 GB	123M
Maya1	cuda	16.57s	14.80s	0.53×	0.59×	2.42 GB	6.72 GB	3B
Dia 1.6B-0626	cuda	25.42s	22.82s	0.46×	0.55×	4.42 GB	6.32 GB	1.6B
Sesame CSM-1B	cuda	12.01s	12.45s	0.52×	0.54×	2.38 GB	3.51 GB	1B
ZipVoice 123M (3/5 ok)	cpu	22.96s	13.44s	0.27×	0.45×	35.45 GB	—	123M
VoxCPM2 2B	cpu	15.35s	14.85s	0.46×	0.45×	10.14 GB	—	2B
Chatterbox	cpu	14.47s	14.10s	0.40×	0.43×	4.24 GB	—	1.2B
OmniVoice	cpu	16.45s	15.98s	0.38×	0.39×	3.05 GB	—	~1B
OuteTTS 1.0 1B	cuda	25.01s	24.27s	0.32×	0.33×	2.59 GB	3.67 GB	1B
Miso TTS 8B	cuda	36.25s	32.11s	0.30×	0.31×	2.01 GB	16.53 GB	8.2B
Magpie-TTS	cpu	37.17s	33.72s	0.29×	0.30×	6.10 GB	—	357M
Mars5-TTS	cpu	31.19s	30.69s	0.22×	0.24×	4.03 GB	—	1.2B
Mars5-TTS	cuda	31.20s	31.06s	0.23×	0.23×	2.25 GB	6.81 GB	1.2B
VibeVoice 1.5B	cpu	39.59s	45.31s	0.19×	0.20×	11.62 GB	—	1.5B
Qwen3-TTS 1.7B Base	cpu	34.81s	30.07s	0.18×	0.19×	10.40 GB	—	1.7B
Fish Speech 1.5	cpu	45.73s	45.34s	0.17×	0.17×	4.46 GB	—	~500M
Parler-TTS Mini v1	cpu	63.39s	62.77s	0.14×	0.14×	4.31 GB	—	878M
Zonos v0.1	cpu	62.12s	60.57s	0.12×	0.12×	7.39 GB	—	1.6B
Sesame CSM-1B	cpu	50.90s	57.54s	0.11×	0.12×	5.77 GB	—	1B
F5-TTS v1	cpu	58.77s	60.21s	0.07×	0.07×	2.59 GB	—	330M
Maya1 (2/4 ok)	cpu	66.31s	73.66s	0.07×	0.07×	7.33 GB	—	3B
LuxTTS	cpu	LuxTTS install failed (piper-phonemize has no Windows wheels)

Model

Device

TTFA cold

TTFA warm

RTF cold

RTF warm

Peak RAM

Peak VRAM

Size

Kokoro

cuda

895ms

67ms

8.32×

103.68×

2.43 GB

925 MB

82M

MeloTTS

cuda

1.61s

117ms

4.60×

60.41×

2.89 GB

1.16 GB

~52M

Piper

cpu

160ms

107ms

38.94×

58.83×

470 MB

—

~25MB

StyleTTS 2

cpu

1.74s

251ms

4.93×

33.76×

2.61 GB

—

~148M

StyleTTS 2

cuda

1.76s

265ms

4.80×

32.15×

2.60 GB

1.49 GB

~148M

OpenVoice v2

cuda

1.85s

361ms

3.65×

16.93×

2.65 GB

1.35 GB

~100M

Kokoro

cpu

609ms

532ms

12.13×

14.38×

1.82 GB

—

82M

Supertonic 3

cpu

744ms

741ms

9.86×

9.92×

570 MB

—

99M

MeloTTS

cpu

1.88s

794ms

3.68×

9.40×

2.48 GB

—

~52M

OmniVoice

cuda

1.12s

757ms

6.44×

9.30×

2.07 GB

2.16 GB

~1B

LongCat-AudioDiT 1B

cuda

1.76s

1.19s

5.10×

6.69×

2.22 GB

5.71 GB

1.42B

KittenTTS Nano 0.1

cpu

1.22s

1.20s

6.40×

6.39×

338 MB

—

<100M

F5-TTS v1

cuda

1.31s

845ms

3.47×

5.32×

2.67 GB

802 MB

330M

Coqui XTTS-v2

cuda

2.05s

1.87s

3.63×

4.75×

2.10 GB

2.14 GB

750M

Chatterbox Turbo

cuda

2.39s

1.62s

2.80×

4.28×

2.44 GB

3.01 GB

744M

LongCat-AudioDiT 3.5B

cuda

2.27s

1.80s

3.41×

4.24×

2.22 GB

14.96 GB

3.83B

OpenVoice v2

cpu

2.83s

1.48s

2.10×

4.10×

2.78 GB

—

~100M

Pocket-TTS

cpu

147ms

123ms

3.99×

4.06×

1.95 GB

—

100M

Soprano 1.1 80M

cuda

1.74s

1.77s

3.77×

3.76×

2.12 GB

326 MB

80M

Qwen3-TTS 1.7B (CUDA-graph)

cuda

6.51s

1.60s

0.90×

3.76×

2.48 GB

4.89 GB

1.7B

Echo-TTS

cuda

2.83s

2.15s

2.63×

3.44×

1.94 GB

9.38 GB

2.8B

Soprano 1.1 80M

cpu

1.97s

2.00s

3.40×

1.34 GB

—

80M

NeuTTS Nano

cuda

678ms

258ms

2.19×

2.76×

3.24 GB

3.26 GB

229M

DramaBox

cuda

5.14s

3.78s

1.93×

2.58×

2.36 GB

17.39 GB

3.3B

VibeVoice Realtime 0.5B

cuda

3.80s

3.77s

2.24×

2.39×

1.88 GB

2.62 GB

0.5B

Chatterbox

cuda

3.35s

2.61s

1.66×

2.24×

2.80 GB

3.24 GB

1.2B

NeuTTS Nano

cpu

698ms

303ms

1.73×

2.00×

5.03 GB

—

229M

Magpie-TTS

cuda

5.48s

4.48s

1.53×

1.93×

3.54 GB

5.60 GB

357M

NeuTTS Air

cuda

1.15s

417ms

1.34×

1.62×

3.60 GB

3.26 GB

748M

MOSS-TTS-Nano

cuda

5.83s

4.92s

1.36×

1.61×

2.42 GB

777 MB

100M

VibeVoice 1.5B

cuda

4.99s

5.32s

1.44×

1.61×

2.04 GB

5.26 GB

1.5B

VibeVoice 7B

cuda

5.67s

6.17s

1.39×

1.55×

2.05 GB

17.63 GB

MOSS-TTS v1.0

cuda

5.23s

4.74s

1.34×

1.48×

2.08 GB

22.83 GB

VoxCPM2 2B

cuda

5.36s

5.19s

1.35×

6.18 GB

5.65 GB

IndexTTS-2

cpu

6.48s

5.42s

1.09×

1.31×

5.56 GB

—

1.5B

NeuTTS Air

cpu

1.20s

471ms

1.16×

1.29×

5.37 GB

—

748M

LFM2.5-Audio 1.5B

cpu

391ms

84ms

1.02×

1.26×

2.25 GB

—

1.5B

MOSS-TTS-Nano

cpu

6.38s

5.64s

1.08×

1.21×

3.14 GB

—

100M

LFM2.5-Audio 1.5B

cuda

410ms

91ms

0.99×

1.18×

2.25 GB

3.18 GB

1.5B

IndexTTS-2

cuda

7.20s

6.03s

0.93×

1.11×

5.88 GB

7.60 GB

1.5B

Parler-TTS Mini v1

cuda

8.29s

8.14s

0.96×

1.05×

3.19 GB

2.63 GB

878M

MOSS-TTS v1.5

cuda

7.66s

6.86s

0.88×

0.99×

2.38 GB

22.83 GB

Fish Speech 1.5

cuda

8.53s

7.84s

0.81×

0.94×

3.83 GB

1.80 GB

~500M

Coqui XTTS-v2

cpu

9.92s

9.73s

0.87×

0.88×

3.23 GB

—

750M

Chatterbox Turbo

cpu

10.65s

9.92s

0.70×

0.73×

4.16 GB

—

744M

Zonos v0.1

cuda

10.88s

10.38s

0.66×

0.71×

6.51 GB

4.48 GB

1.6B

Qwen3-TTS 1.7B Base

cuda

10.77s

8.95s

0.55×

0.70×

2.42 GB

4.64 GB

1.7B

VibeVoice Realtime 0.5B

cpu

14.49s

13.35s

0.63×

0.66×

5.89 GB

—

0.5B

ZipVoice 123M (4/5 ok)

cuda

68.83s

86.51s

0.35×

0.60×

25.87 GB

53.16 GB

123M

Maya1

cuda

16.57s

14.80s

0.53×

0.59×

2.42 GB

6.72 GB

Dia 1.6B-0626

cuda

25.42s

22.82s

0.46×

0.55×

4.42 GB

6.32 GB

1.6B

Sesame CSM-1B

cuda

12.01s

12.45s

0.52×

0.54×

2.38 GB

3.51 GB

ZipVoice 123M (3/5 ok)

cpu

22.96s

13.44s

0.27×

0.45×

35.45 GB

—

123M

VoxCPM2 2B

cpu

15.35s

14.85s

0.46×

0.45×

10.14 GB

—

Chatterbox

cpu

14.47s

14.10s

0.40×

0.43×

4.24 GB

—

1.2B

OmniVoice

cpu

16.45s

15.98s

0.38×

0.39×

3.05 GB

—

~1B

OuteTTS 1.0 1B

cuda

25.01s

24.27s

0.32×

0.33×

2.59 GB

3.67 GB

Miso TTS 8B

cuda

36.25s

32.11s

0.30×

0.31×

2.01 GB

16.53 GB

8.2B

Magpie-TTS

cpu

37.17s

33.72s

0.29×

0.30×

6.10 GB

—

357M

Mars5-TTS

cpu

31.19s

30.69s

0.22×

0.24×

4.03 GB

—

1.2B

Mars5-TTS

cuda

31.20s

31.06s

0.23×

2.25 GB

6.81 GB

1.2B

VibeVoice 1.5B

cpu

39.59s

45.31s

0.19×

0.20×

11.62 GB

—

1.5B

Qwen3-TTS 1.7B Base

cpu

34.81s

30.07s

0.18×

0.19×

10.40 GB

—

1.7B

Fish Speech 1.5

cpu

45.73s

45.34s

0.17×

4.46 GB

—

~500M

Parler-TTS Mini v1

cpu

63.39s

62.77s

0.14×

4.31 GB

—

878M

Zonos v0.1

cpu

62.12s

60.57s

0.12×

7.39 GB

—

1.6B

Sesame CSM-1B

cpu

50.90s

57.54s

0.11×

0.12×

5.77 GB

—

F5-TTS v1

cpu

58.77s

60.21s

0.07×

2.59 GB

—

330M

Maya1 (2/4 ok)

cpu

66.31s

73.66s

0.07×

7.33 GB

—

LuxTTS

cpu

LuxTTS install failed (piper-phonemize has no Windows wheels)