click any column header to sort
← SpeedSpeedSamples → · all runs

TTS Bench — Samples — 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
5 prompt(s) · one section per prompt · all models ranked by warm TTFA (fastest first) within each
Each prompt section shows every model's audio output, ordered by warm TTFA (fastest first). Click any audio player to hear that model's rendering.

Prompt 1

[en]"Open the browser and read my email."
Rank Model Device TTFA warm Audio
1 sanoTTS Heart-Nano cpu 25ms
2 Vaniq-Edge cuda 31ms
3 Kokoro cuda 47ms
4 Piper cpu 47ms
5 Pocket-TTS cpu 62ms
6 MeloTTS cuda 75ms
7 SupraTTS 0.1 Beta cuda 80ms
8 Vaniq-Edge cpu 82ms
9 LFM2.5-Audio 1.5B cuda 89ms
10 Sopro V2 Turbo (streaming) cuda 107ms
11 Inflect-Micro v2 cuda 127ms
12 Inflect-Nano v2 cuda 128ms
13 sanoTTS Amy cpu 161ms
14 Sopro V2 Turbo cuda 162ms
15 StyleTTS 2 cuda 175ms
16 LFM2.5-Audio 1.5B cpu 189ms
17 Inflect-Micro v2 cpu 200ms
18 Audio8 TTS 0.6B (compiled) cuda 207ms
19 Kokoro cpu 226ms
20 SupraTTS 0.1 Beta cpu 232ms
21 Inflect-Nano v2 cpu 237ms
22 NeuTTS Nano cuda 270ms
23 NeuTTS Nano cpu 303ms
24 OpenVoice v2 cuda 304ms
25 MeloTTS cpu 345ms
26 NeuTTS Air cuda 411ms
27 Supertonic 3 cpu 438ms
28 NeuTTS Air cpu 463ms
29 KittenTTS Nano 0.1 cpu 504ms
30 Sopro V2 Turbo (streaming) cpu 507ms
31 AuK-Flash cuda 514ms
32 Scylla's Band cpu 535ms
33 Qwen3-TTS 1.7B (CUDA-graph) cuda 557ms
34 Soprano 1.1 80M cuda 557ms
35 OpenVoice v2 cpu 595ms
36 Coqui XTTS-v2 cuda 614ms
37 Soprano 1.1 80M cpu 661ms
38 StyleTTS 2 cpu 688ms
39 Chatterbox Turbo cuda 725ms
40 F5-TTS v1 cuda 778ms
41 OmniVoice cuda 887ms
42 ZipVoice 123M cuda 1.20s
43 Chatterbox cuda 1.22s
44 LongCat-AudioDiT 1B cuda 1.24s
45 Sopro V2 Turbo cpu 1.25s
46 FireRedTTS3 cuda 1.29s
47 VibeVoice Realtime 0.5B cuda 1.35s
48 Magpie-TTS cuda 1.61s
49 LongCat-AudioDiT 3.5B cuda 1.76s
50 VibeVoice 1.5B cuda 1.85s
51 Echo-TTS cuda 2.11s
52 VoxCPM2 2B cuda 2.48s
53 MOSS-TTS-Nano cuda 2.57s
54 Parler-TTS Mini v1 cuda 2.58s
55 MOSS-TTS v1.0 cuda 2.72s
56 IndexTTS-2 cuda 2.74s
57 Coqui XTTS-v2 cpu 2.75s
58 Qwen3-TTS 1.7B Base cuda 2.78s
59 MOSS-TTS-Nano cpu 2.99s
60 DramaBox cuda 2.99s
61 Fish Speech 1.5 cuda 3.06s
62 AuK cuda 3.36s
63 Zonos v0.1 cuda 3.50s
64 Chatterbox Turbo cpu 3.62s
65 Sesame CSM-1B cuda 3.73s
66 Audio8 TTS Preview 0.6B cuda 3.96s
67 VibeVoice Realtime 0.5B cpu 4.74s
68 Maya1 cuda 4.75s
69 Dia 1.6B-0626 cuda 4.75s
70 Audio8 TTS Preview 0.1B cuda 4.87s
71 Qwen3-TTS 0.6B CustomVoice cuda 4.94s
72 Chatterbox cpu 5.32s
73 Miso TTS 8B cuda 5.48s
74 MOSS-TTS v1.5 cuda 5.62s
75 Magpie-TTS cpu 7.12s
76 OmniVoice cpu 7.50s
77 VibeVoice 7B cuda 7.64s
78 VoxCPM2 2B cpu 8.44s
79 OuteTTS 1.0 1B cuda 10.20s
80 ZipVoice 123M cpu 10.63s
81 Qwen3-TTS 1.7B Base cpu 11.69s
82 Qwen3-TTS 0.6B CustomVoice cpu 13.53s
83 Audio8 TTS Preview 0.6B cpu 13.96s
84 Audio8 TTS Preview 0.1B cpu 14.47s
85 Fish Speech 1.5 cpu 15.07s
86 Parler-TTS Mini v1 cpu 16.48s
87 Mars5-TTS cuda 19.75s
88 Zonos v0.1 cpu 19.96s
89 VibeVoice 1.5B cpu 20.22s
90 Sesame CSM-1B cpu 20.89s
91 IndexTTS-2 cpu 39.34s
92 F5-TTS v1 cpu 45.35s
93 Maya1 cpu 49.55s

Prompt 2

[en]"I'll start a new git branch, push the changes, and open a pull request when the tests pass."
Rank Model Device TTFA warm Audio
1 Vaniq-Edge cuda 27ms
2 Kokoro cuda 59ms
3 sanoTTS Heart-Nano cpu 64ms
4 Piper cpu 89ms
5 Pocket-TTS cpu 90ms
6 LFM2.5-Audio 1.5B cuda 103ms
7 Sopro V2 Turbo (streaming) cuda 107ms
8 MeloTTS cuda 107ms
9 Inflect-Nano v2 cuda 117ms
10 Inflect-Micro v2 cuda 124ms
11 Vaniq-Edge cpu 170ms
12 StyleTTS 2 cuda 188ms
13 LFM2.5-Audio 1.5B cpu 231ms
14 Inflect-Nano v2 cpu 247ms
15 NeuTTS Nano cuda 268ms
16 SupraTTS 0.1 Beta cuda 277ms
17 Sopro V2 Turbo cuda 311ms
18 NeuTTS Nano cpu 317ms
19 OpenVoice v2 cuda 353ms
20 Inflect-Micro v2 cpu 361ms
21 sanoTTS Amy cpu 372ms
22 Kokoro cpu 380ms
23 NeuTTS Air cuda 420ms
24 Audio8 TTS 0.6B (compiled) cuda 434ms
25 AuK-Flash cuda 482ms
26 NeuTTS Air cpu 506ms
27 Sopro V2 Turbo (streaming) cpu 518ms
28 Scylla's Band cpu 556ms
29 Supertonic 3 cpu 586ms
30 MeloTTS cpu 656ms
31 SupraTTS 0.1 Beta cpu 679ms
32 F5-TTS v1 cuda 735ms
33 OmniVoice cuda 773ms
34 KittenTTS Nano 0.1 cpu 843ms
35 LongCat-AudioDiT 1B cuda 971ms
36 Coqui XTTS-v2 cuda 1.15s
37 OpenVoice v2 cpu 1.18s
38 Qwen3-TTS 1.7B (CUDA-graph) cuda 1.30s
39 Chatterbox Turbo cuda 1.30s
40 Soprano 1.1 80M cuda 1.33s
41 StyleTTS 2 cpu 1.39s
42 Soprano 1.1 80M cpu 1.43s
43 LongCat-AudioDiT 3.5B cuda 1.49s
44 Sopro V2 Turbo cpu 2.19s
45 Echo-TTS cuda 2.20s
46 Chatterbox cuda 2.24s
47 VibeVoice Realtime 0.5B cuda 2.64s
48 DramaBox cuda 3.42s
49 Magpie-TTS cuda 3.45s
50 VibeVoice 1.5B cuda 3.47s
51 VoxCPM2 2B cuda 3.59s
52 AuK cuda 3.61s
53 VibeVoice 7B cuda 3.78s
54 MOSS-TTS v1.0 cuda 3.94s
55 IndexTTS-2 cuda 4.16s
56 MOSS-TTS-Nano cpu 4.47s
57 MOSS-TTS-Nano cuda 5.23s
58 MOSS-TTS v1.5 cuda 5.89s
59 Parler-TTS Mini v1 cuda 6.04s
60 FireRedTTS3 cuda 6.41s
61 Qwen3-TTS 1.7B Base cuda 6.47s
62 Coqui XTTS-v2 cpu 6.65s
63 Fish Speech 1.5 cuda 7.07s
64 Zonos v0.1 cuda 7.78s
65 Chatterbox Turbo cpu 8.06s
66 Audio8 TTS Preview 0.6B cuda 8.70s
67 VibeVoice Realtime 0.5B cpu 9.36s
68 Qwen3-TTS 0.6B CustomVoice cuda 10.10s
69 Chatterbox cpu 10.32s
70 Sesame CSM-1B cuda 10.43s
71 Maya1 cuda 10.54s
72 VoxCPM2 2B cpu 11.56s
73 OmniVoice cpu 13.09s
74 Audio8 TTS Preview 0.1B cuda 14.39s
75 OuteTTS 1.0 1B cuda 20.98s
76 Magpie-TTS cpu 21.02s
77 Miso TTS 8B cuda 22.52s
78 Qwen3-TTS 1.7B Base cpu 22.87s
79 Audio8 TTS Preview 0.1B cpu 24.24s
80 Qwen3-TTS 0.6B CustomVoice cpu 24.54s
81 Mars5-TTS cuda 25.09s
82 VibeVoice 1.5B cpu 29.51s
83 Fish Speech 1.5 cpu 31.14s
84 Audio8 TTS Preview 0.6B cpu 31.27s
85 Parler-TTS Mini v1 cpu 42.46s
86 Zonos v0.1 cpu 45.49s
87 Dia 1.6B-0626 cuda 51.96s
88 F5-TTS v1 cpu 58.82s
89 IndexTTS-2 cpu 61.67s
90 Sesame CSM-1B cpu 66.45s
91 Maya1 cpu 97.77s
92 ZipVoice 123M cuda 137.13s

Prompt 3

[en]"The Parakeet TDT zero point six billion parameter model achieves one point six nine percent word error rate on LibriSpeech test-clean, beating Whisper Large V3 at two point seven percent while running at over two thousand times realtime on a single GPU."
Rank Model Device TTFA warm Audio
1 Vaniq-Edge cuda 33ms
2 LFM2.5-Audio 1.5B cuda 79ms
3 Pocket-TTS cpu 102ms
4 Sopro V2 Turbo (streaming) cuda 103ms
5 Kokoro cuda 116ms
6 Inflect-Micro v2 cuda 136ms
7 SupraTTS 0.1 Beta cuda 149ms
8 MeloTTS cuda 153ms
9 Inflect-Nano v2 cuda 167ms
10 LFM2.5-Audio 1.5B cpu 218ms
11 Piper cpu 239ms
12 NeuTTS Nano cuda 265ms
13 Sopro V2 Turbo (streaming) cpu 269ms
14 NeuTTS Nano cpu 322ms
15 StyleTTS 2 cuda 374ms
16 NeuTTS Air cuda 423ms
17 NeuTTS Air cpu 458ms
18 OpenVoice v2 cuda 485ms
19 Vaniq-Edge cpu 568ms
20 Inflect-Nano v2 cpu 593ms
21 AuK-Flash cuda 674ms
22 OmniVoice cuda 714ms
23 Sopro V2 Turbo cuda 870ms
24 Inflect-Micro v2 cpu 877ms
25 sanoTTS Amy cpu 1.03s
26 F5-TTS v1 cuda 1.03s
27 Audio8 TTS 0.6B (compiled) cuda 1.18s
28 Scylla's Band cpu 1.27s
29 Kokoro cpu 1.27s
30 LongCat-AudioDiT 1B cuda 1.29s
31 Supertonic 3 cpu 1.45s
32 SupraTTS 0.1 Beta cpu 1.53s
33 MeloTTS cpu 1.76s
34 Echo-TTS cuda 2.20s
35 LongCat-AudioDiT 3.5B cuda 2.20s
36 KittenTTS Nano 0.1 cpu 2.46s
37 OpenVoice v2 cpu 3.11s
38 Chatterbox Turbo cuda 3.12s
39 Qwen3-TTS 1.7B (CUDA-graph) cuda 3.30s
40 Soprano 1.1 80M cuda 3.70s
41 StyleTTS 2 cpu 3.93s
42 Soprano 1.1 80M cpu 4.27s
43 Chatterbox cuda 4.68s
44 Coqui XTTS-v2 cuda 4.82s
45 AuK cuda 4.99s
46 DramaBox cuda 5.19s
47 Sopro V2 Turbo cpu 5.27s
48 VibeVoice Realtime 0.5B cuda 7.86s
49 MOSS-TTS v1.0 cuda 8.40s
50 VibeVoice 7B cuda 8.77s
51 Magpie-TTS cuda 9.64s
52 MOSS-TTS-Nano cuda 10.45s
53 VibeVoice 1.5B cuda 10.63s
54 VoxCPM2 2B cuda 11.06s
55 MOSS-TTS-Nano cpu 11.83s
56 MOSS-TTS v1.5 cuda 11.88s
57 IndexTTS-2 cuda 12.58s
58 FireRedTTS3 cuda 17.24s
59 Fish Speech 1.5 cuda 17.39s
60 Parler-TTS Mini v1 cuda 18.07s
61 Sesame CSM-1B cuda 18.23s
62 Chatterbox Turbo cpu 19.26s
63 Qwen3-TTS 1.7B Base cuda 19.42s
64 Dia 1.6B-0626 cuda 22.29s
65 Audio8 TTS Preview 0.6B cuda 24.09s
66 Coqui XTTS-v2 cpu 25.28s
67 Zonos v0.1 cuda 25.58s
68 VibeVoice Realtime 0.5B cpu 26.34s
69 Chatterbox cpu 28.27s
70 Maya1 cuda 29.31s
71 Qwen3-TTS 0.6B CustomVoice cuda 30.99s
72 VoxCPM2 2B cpu 32.23s
73 OmniVoice cpu 34.51s
74 Audio8 TTS Preview 0.1B cuda 42.41s
75 OuteTTS 1.0 1B cuda 50.39s
76 Mars5-TTS cuda 51.19s
77 Qwen3-TTS 1.7B Base cpu 63.06s
78 Audio8 TTS Preview 0.1B cpu 74.04s
79 Miso TTS 8B cuda 75.59s
80 F5-TTS v1 cpu 77.80s
81 Sesame CSM-1B cpu 78.74s
82 Qwen3-TTS 0.6B CustomVoice cpu 79.10s
83 VibeVoice 1.5B cpu 85.74s
84 Magpie-TTS cpu 87.47s
85 Fish Speech 1.5 cpu 96.22s
86 Audio8 TTS Preview 0.6B cpu 101.71s
87 Parler-TTS Mini v1 cpu 140.05s
88 IndexTTS-2 cpu 144.93s
89 Zonos v0.1 cpu 156.70s

Prompt 4

[en]"Run pytest tests slash test underscore voice dot py with verbose flag and capture flag set to no."
Rank Model Device TTFA warm Audio
1 Vaniq-Edge cuda 27ms
2 sanoTTS Heart-Nano cpu 60ms
3 Kokoro cuda 65ms
4 SupraTTS 0.1 Beta cuda 83ms
5 Pocket-TTS cpu 90ms
6 LFM2.5-Audio 1.5B cuda 91ms
7 Piper cpu 93ms
8 MeloTTS cuda 108ms
9 Inflect-Nano v2 cuda 113ms
10 Inflect-Micro v2 cuda 114ms
11 Sopro V2 Turbo (streaming) cuda 115ms
12 Vaniq-Edge cpu 180ms
13 LFM2.5-Audio 1.5B cpu 218ms
14 Inflect-Nano v2 cpu 255ms
15 NeuTTS Nano cuda 258ms
16 Sopro V2 Turbo (streaming) cpu 273ms
17 NeuTTS Nano cpu 297ms
18 StyleTTS 2 cuda 323ms
19 Inflect-Micro v2 cpu 370ms
20 sanoTTS Amy cpu 379ms
21 OpenVoice v2 cuda 387ms
22 Sopro V2 Turbo cuda 408ms
23 NeuTTS Air cuda 417ms
24 Kokoro cpu 452ms
25 NeuTTS Air cpu 456ms
26 SupraTTS 0.1 Beta cpu 491ms
27 AuK-Flash cuda 532ms
28 Supertonic 3 cpu 663ms
29 Audio8 TTS 0.6B (compiled) cuda 673ms
30 OmniVoice cuda 694ms
31 MeloTTS cpu 743ms
32 Scylla's Band cpu 814ms
33 F5-TTS v1 cuda 836ms
34 KittenTTS Nano 0.1 cpu 1.01s
35 LongCat-AudioDiT 1B cuda 1.26s
36 Chatterbox Turbo cuda 1.33s
37 OpenVoice v2 cpu 1.39s
38 Soprano 1.1 80M cuda 1.49s
39 StyleTTS 2 cpu 1.57s
40 Soprano 1.1 80M cpu 1.62s
41 Qwen3-TTS 1.7B (CUDA-graph) cuda 1.66s
42 LongCat-AudioDiT 3.5B cuda 1.74s
43 Coqui XTTS-v2 cuda 1.93s
44 Echo-TTS cuda 2.09s
45 Chatterbox cuda 2.31s
46 Sopro V2 Turbo cpu 2.92s
47 AuK cuda 2.92s
48 VibeVoice Realtime 0.5B cuda 3.23s
49 DramaBox cuda 3.52s
50 MOSS-TTS-Nano cuda 3.67s
51 Magpie-TTS cuda 4.37s
52 VibeVoice 7B cuda 4.48s
53 IndexTTS-2 cuda 4.64s
54 MOSS-TTS v1.0 cuda 4.90s
55 MOSS-TTS-Nano cpu 5.00s
56 VibeVoice 1.5B cuda 5.30s
57 Parler-TTS Mini v1 cuda 5.87s
58 VoxCPM2 2B cuda 6.05s
59 Fish Speech 1.5 cuda 6.90s
60 MOSS-TTS v1.5 cuda 7.04s
61 Chatterbox Turbo cpu 8.76s
62 FireRedTTS3 cuda 9.05s
63 Zonos v0.1 cuda 9.44s
64 Coqui XTTS-v2 cpu 9.48s
65 Qwen3-TTS 1.7B Base cuda 9.49s
66 Audio8 TTS Preview 0.6B cuda 10.83s
67 Dia 1.6B-0626 cuda 12.27s
68 Chatterbox cpu 12.46s
69 VibeVoice Realtime 0.5B cpu 12.96s
70 OmniVoice cpu 13.20s
71 VoxCPM2 2B cpu 13.91s
72 Maya1 cuda 14.58s
73 Qwen3-TTS 0.6B CustomVoice cuda 15.33s
74 ZipVoice 123M cpu 16.26s
75 Sesame CSM-1B cuda 17.41s
76 Audio8 TTS Preview 0.1B cuda 18.41s
77 OuteTTS 1.0 1B cuda 23.43s
78 Miso TTS 8B cuda 24.87s
79 Mars5-TTS cuda 28.21s
80 Magpie-TTS cpu 29.11s
81 Qwen3-TTS 1.7B Base cpu 30.68s
82 Audio8 TTS Preview 0.1B cpu 34.22s
83 Qwen3-TTS 0.6B CustomVoice cpu 35.06s
84 Fish Speech 1.5 cpu 42.69s
85 VibeVoice 1.5B cpu 45.79s
86 Audio8 TTS Preview 0.6B cpu 47.85s
87 Zonos v0.1 cpu 49.69s
88 Parler-TTS Mini v1 cpu 52.09s
89 F5-TTS v1 cpu 58.87s
90 Sesame CSM-1B cpu 64.06s
91 IndexTTS-2 cpu 75.35s
92 ZipVoice 123M cuda 142.94s

Prompt 5

[fr]"Bonjour, je m'appelle Cicero et je vais vous aider avec votre code aujourd'hui."
Rank Model Device TTFA warm Audio
1 Kokoro cuda 50ms
2 Piper cpu 68ms
3 Sopro V2 Turbo (streaming) cuda 113ms
4 MeloTTS cuda 139ms
5 NeuTTS Nano cuda 230ms
6 Sopro V2 Turbo cuda 247ms
7 Pocket-TTS cpu 253ms
8 OpenVoice v2 cuda 276ms
9 NeuTTS Nano cpu 277ms
10 Kokoro cpu 334ms
11 Audio8 TTS 0.6B (compiled) cuda 398ms
12 AuK-Flash cuda 457ms
13 MeloTTS cpu 464ms
14 Supertonic 3 cpu 565ms
15 OmniVoice cuda 719ms
16 Sopro V2 Turbo (streaming) cpu 805ms
17 Coqui XTTS-v2 cuda 859ms
18 OpenVoice v2 cpu 1.13s
19 Qwen3-TTS 1.7B (CUDA-graph) cuda 1.18s
20 Sopro V2 Turbo cpu 2.05s
21 MOSS-TTS-Nano cuda 2.67s
22 VoxCPM2 2B cuda 2.77s
23 AuK cuda 3.21s
24 Magpie-TTS cuda 3.34s
25 MOSS-TTS v1.0 cuda 3.74s
26 MOSS-TTS v1.5 cuda 3.87s
27 MOSS-TTS-Nano cpu 3.89s
28 Coqui XTTS-v2 cpu 4.47s
29 Fish Speech 1.5 cuda 4.77s
30 Zonos v0.1 cuda 5.59s
31 Audio8 TTS Preview 0.6B cuda 6.49s
32 Qwen3-TTS 1.7B Base cuda 6.57s
33 Qwen3-TTS 0.6B CustomVoice cuda 8.04s
34 VoxCPM2 2B cpu 8.08s
35 OmniVoice cpu 11.60s
36 Audio8 TTS Preview 0.1B cuda 11.91s
37 ZipVoice 123M cpu 13.44s
38 OuteTTS 1.0 1B cuda 16.34s
39 Audio8 TTS Preview 0.1B cpu 16.73s
40 Qwen3-TTS 0.6B CustomVoice cpu 19.71s
41 Qwen3-TTS 1.7B Base cpu 22.03s
42 Magpie-TTS cpu 23.89s
43 Audio8 TTS Preview 0.6B cpu 26.79s
44 Zonos v0.1 cpu 30.99s
45 Fish Speech 1.5 cpu 41.60s
46 ZipVoice 123M cuda 64.78s