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

TTS Bench — Samples — linux-default

Rig: linux-3090 — AMD Ryzen 9 5900XT 16-Core Processor · NVIDIA GeForce RTX 3090 24GB · 63 GB RAM · Linux 6.8.0-117-generic
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 Kokoro cuda 47ms
2 Piper cpu 62ms
3 Pocket-TTS cpu 85ms
4 StyleTTS 2 cuda 142ms
5 StyleTTS 2 cpu 144ms
6 LuxTTS cuda 194ms
7 OpenVoice v2 cuda 248ms
8 NeuTTS Nano cuda 315ms
9 Kokoro cpu 361ms
10 NeuTTS Nano cpu 392ms
11 Soprano 1.1 80M cuda 397ms
12 MiraTTS cuda 451ms
13 NeuTTS Air cuda 486ms
14 Chatterbox Turbo cuda 521ms
15 NeuTTS Air cpu 546ms
16 Coqui XTTS-v2 cuda 557ms
17 Supertonic 3 cpu 596ms
18 Soprano 1.1 80M cpu 662ms
19 Qwen3-TTS 1.7B (CUDA-graph) cuda 679ms
20 OmniVoice cuda 713ms
21 KittenTTS Nano 0.1 cpu 736ms
22 Higgs Audio v3 TTS cuda 820ms
23 longcat_1b cuda 856ms
24 OpenVoice v2 cpu 1.04s
25 longcat_3p5b cuda 1.05s
26 Chatterbox cuda 1.06s
27 VibeVoice Realtime 0.5B cuda 1.08s
28 LuxTTS cpu 1.10s
29 F5-TTS v1 cuda 1.11s
30 Magpie-TTS cuda 1.42s
31 VibeVoice 1.5B cuda 1.46s
32 MOSS-TTS-Nano cuda 1.66s
33 Fish Speech 1.5 cuda 2.03s
34 Qwen3-TTS 1.7B Base cuda 2.08s
35 MOSS-TTS v1.0 cuda 2.29s
36 Zonos v0.1 cuda 2.36s
37 VoxCPM2 2B cuda 2.36s
38 MOSS-TTS v1.5 cuda 2.48s
39 MetaVoice-1B cuda 2.52s
40 IndexTTS-2 cpu 2.65s
41 IndexTTS-2 cuda 2.88s
42 Step-Audio-EditX cuda 2.92s
43 Coqui XTTS-v2 cpu 3.07s
44 Sesame CSM-1B cuda 3.24s
45 MOSS-TTS-Nano cpu 3.38s
46 dots.tts (soar) cuda 3.77s
47 Chatterbox Turbo cpu 3.79s
48 VibeVoice 7B cuda 3.88s
49 Dia 1.6B-0626 cuda 4.12s
50 Maya1 cuda 4.28s
51 Echo-TTS cuda 4.33s
52 VibeVoice Realtime 0.5B cpu 5.46s
53 Miso TTS 8B cuda 5.75s
54 DramaBox cuda 5.78s
55 Chatterbox cpu 6.02s
56 ZipVoice 123M cpu 7.14s
57 OuteTTS 1.0 1B cuda 7.24s
58 OmniVoice cpu 7.73s
59 Magpie-TTS cpu 8.50s
60 Fish Speech S2-Pro cuda 11.36s
61 Qwen3-TTS 1.7B Base cpu 13.70s
62 VoxCPM2 2B cpu 13.71s
63 Sesame CSM-1B cpu 14.84s
64 Zonos v0.1 cpu 18.49s
65 VibeVoice 1.5B cpu 19.39s
66 Fish Speech 1.5 cpu 21.81s
67 Maya1 cpu 35.94s
68 Mars5-TTS cpu 37.79s
69 Mars5-TTS cuda 37.89s
70 F5-TTS v1 cpu 38.91s
71 OuteTTS 1.0 1B cpu 64.24s

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 Kokoro cuda 63ms
2 Pocket-TTS cpu 110ms
3 Piper cpu 150ms
4 StyleTTS 2 cuda 203ms
5 LuxTTS cuda 205ms
6 StyleTTS 2 cpu 212ms
7 OpenVoice v2 cuda 287ms
8 NeuTTS Nano cuda 315ms
9 NeuTTS Nano cpu 390ms
10 NeuTTS Air cuda 462ms
11 NeuTTS Air cpu 547ms
12 OmniVoice cuda 678ms
13 longcat_1b cuda 787ms
14 MiraTTS cuda 806ms
15 Kokoro cpu 824ms
16 Supertonic 3 cpu 1.00s
17 Soprano 1.1 80M cuda 1.01s
18 Chatterbox Turbo cuda 1.12s
19 Coqui XTTS-v2 cuda 1.17s
20 F5-TTS v1 cuda 1.34s
21 LuxTTS cpu 1.57s
22 Qwen3-TTS 1.7B (CUDA-graph) cuda 1.59s
23 longcat_3p5b cuda 1.66s
24 Higgs Audio v3 TTS cuda 1.71s
25 Soprano 1.1 80M cpu 1.75s
26 Chatterbox cuda 1.90s
27 KittenTTS Nano 0.1 cpu 1.91s
28 OpenVoice v2 cpu 2.24s
29 VibeVoice Realtime 0.5B cuda 2.24s
30 VoxCPM2 2B cuda 2.43s
31 VibeVoice 1.5B cuda 2.51s
32 MOSS-TTS-Nano cuda 2.86s
33 MOSS-TTS v1.5 cuda 3.11s
34 Magpie-TTS cuda 3.12s
35 MOSS-TTS v1.0 cuda 3.19s
36 IndexTTS-2 cpu 3.84s
37 Step-Audio-EditX cuda 4.23s
38 IndexTTS-2 cuda 4.27s
39 Fish Speech 1.5 cuda 4.27s
40 Echo-TTS cuda 4.30s
41 MOSS-TTS-Nano cpu 5.18s
42 MetaVoice-1B cuda 5.34s
43 dots.tts (soar) cuda 5.39s
44 Qwen3-TTS 1.7B Base cuda 5.54s
45 Zonos v0.1 cuda 5.72s
46 Sesame CSM-1B cuda 6.62s
47 Chatterbox Turbo cpu 6.65s
48 DramaBox cuda 7.05s
49 Coqui XTTS-v2 cpu 7.15s
50 Maya1 cuda 8.17s
51 VibeVoice 7B cuda 9.32s
52 VibeVoice Realtime 0.5B cpu 11.00s
53 Chatterbox cpu 12.08s
54 OmniVoice cpu 12.82s
55 ZipVoice 123M cpu 13.71s
56 OuteTTS 1.0 1B cuda 15.09s
57 Miso TTS 8B cuda 15.38s
58 Dia 1.6B-0626 cuda 21.78s
59 VoxCPM2 2B cpu 21.93s
60 Magpie-TTS cpu 23.70s
61 Qwen3-TTS 1.7B Base cpu 26.55s
62 Fish Speech S2-Pro cuda 26.67s
63 VibeVoice 1.5B cpu 31.43s
64 Zonos v0.1 cpu 45.78s
65 Mars5-TTS cuda 47.97s
66 Mars5-TTS cpu 47.98s
67 F5-TTS v1 cpu 49.20s
68 Sesame CSM-1B cpu 61.47s
69 Fish Speech 1.5 cpu 68.38s
70 Maya1 cpu 82.85s
71 OuteTTS 1.0 1B cpu 122.25s

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 Kokoro cuda 133ms
2 Pocket-TTS cpu 138ms
3 LuxTTS cuda 257ms
4 NeuTTS Nano cuda 348ms
5 StyleTTS 2 cuda 398ms
6 StyleTTS 2 cpu 403ms
7 NeuTTS Nano cpu 419ms
8 Piper cpu 428ms
9 NeuTTS Air cuda 492ms
10 OpenVoice v2 cuda 566ms
11 NeuTTS Air cpu 568ms
12 OmniVoice cuda 1.05s
13 longcat_1b cuda 1.08s
14 F5-TTS v1 cuda 1.94s
15 MiraTTS cuda 2.32s
16 Supertonic 3 cpu 2.67s
17 longcat_3p5b cuda 2.98s
18 Soprano 1.1 80M cuda 2.99s
19 Chatterbox Turbo cuda 3.19s
20 Kokoro cpu 3.30s
21 LuxTTS cpu 3.33s
22 KittenTTS Nano 0.1 cpu 3.90s
23 Coqui XTTS-v2 cuda 4.15s
24 Echo-TTS cuda 4.32s
25 Qwen3-TTS 1.7B (CUDA-graph) cuda 4.32s
26 Chatterbox cuda 4.60s
27 Soprano 1.1 80M cpu 4.63s
28 Higgs Audio v3 TTS cuda 5.32s
29 VibeVoice Realtime 0.5B cuda 6.56s
30 MOSS-TTS v1.0 cuda 6.89s
31 VoxCPM2 2B cuda 7.01s
32 MOSS-TTS-Nano cuda 7.14s
33 MOSS-TTS v1.5 cuda 7.42s
34 OpenVoice v2 cpu 7.48s
35 Step-Audio-EditX cuda 9.44s
36 Magpie-TTS cuda 9.82s
37 VibeVoice 1.5B cuda 9.97s
38 IndexTTS-2 cpu 10.97s
39 IndexTTS-2 cuda 11.19s
40 DramaBox cuda 11.52s
41 MOSS-TTS-Nano cpu 12.23s
42 Sesame CSM-1B cuda 12.95s
43 Fish Speech 1.5 cuda 13.38s
44 MetaVoice-1B cuda 14.56s
45 Qwen3-TTS 1.7B Base cuda 14.99s
46 dots.tts (soar) cuda 17.68s
47 Zonos v0.1 cuda 17.80s
48 Dia 1.6B-0626 cuda 18.82s
49 Chatterbox Turbo cpu 22.73s
50 Maya1 cuda 25.34s
51 Coqui XTTS-v2 cpu 25.88s
52 VibeVoice 7B cuda 25.90s
53 OmniVoice cpu 31.64s
54 Chatterbox cpu 31.83s
55 VibeVoice Realtime 0.5B cpu 31.98s
56 OuteTTS 1.0 1B cuda 39.77s
57 Miso TTS 8B cuda 50.14s
58 VoxCPM2 2B cpu 59.59s
59 Qwen3-TTS 1.7B Base cpu 72.20s
60 Fish Speech S2-Pro cuda 81.84s
61 F5-TTS v1 cpu 83.01s
62 Sesame CSM-1B cpu 84.06s
63 VibeVoice 1.5B cpu 95.89s
64 Mars5-TTS cpu 96.12s
65 Mars5-TTS cuda 97.23s
66 Magpie-TTS cpu 106.38s
67 Fish Speech 1.5 cpu 141.83s
68 Zonos v0.1 cpu 164.47s

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 Kokoro cuda 70ms
2 Pocket-TTS cpu 123ms
3 Piper cpu 152ms
4 LuxTTS cuda 206ms
5 StyleTTS 2 cuda 269ms
6 StyleTTS 2 cpu 286ms
7 NeuTTS Nano cuda 323ms
8 OpenVoice v2 cuda 367ms
9 NeuTTS Nano cpu 406ms
10 NeuTTS Air cuda 473ms
11 NeuTTS Air cpu 555ms
12 OmniVoice cuda 672ms
13 longcat_1b cuda 794ms
14 Kokoro cpu 959ms
15 Supertonic 3 cpu 1.12s
16 Soprano 1.1 80M cuda 1.15s
17 Chatterbox Turbo cuda 1.34s
18 F5-TTS v1 cuda 1.34s
19 MiraTTS cuda 1.35s
20 LuxTTS cpu 1.62s
21 longcat_3p5b cuda 1.65s
22 KittenTTS Nano 0.1 cpu 1.71s
23 Soprano 1.1 80M cpu 1.88s
24 Coqui XTTS-v2 cuda 1.99s
25 Qwen3-TTS 1.7B (CUDA-graph) cuda 2.07s
26 Chatterbox cuda 2.26s
27 Higgs Audio v3 TTS cuda 2.29s
28 OpenVoice v2 cpu 2.60s
29 MOSS-TTS-Nano cuda 3.08s
30 VibeVoice Realtime 0.5B cuda 3.22s
31 VoxCPM2 2B cuda 3.59s
32 MOSS-TTS v1.0 cuda 3.93s
33 MOSS-TTS v1.5 cuda 4.12s
34 Magpie-TTS cuda 4.32s
35 Echo-TTS cuda 4.32s
36 VibeVoice 1.5B cuda 4.81s
37 Step-Audio-EditX cuda 5.22s
38 IndexTTS-2 cpu 5.23s
39 IndexTTS-2 cuda 5.47s
40 MOSS-TTS-Nano cpu 5.95s
41 Zonos v0.1 cuda 6.19s
42 Fish Speech 1.5 cuda 6.59s
43 MetaVoice-1B cuda 7.53s
44 DramaBox cuda 7.55s
45 Qwen3-TTS 1.7B Base cuda 7.63s
46 dots.tts (soar) cuda 8.82s
47 Chatterbox Turbo cpu 9.17s
48 Dia 1.6B-0626 cuda 10.60s
49 Maya1 cuda 10.95s
50 Sesame CSM-1B cuda 11.84s
51 OmniVoice cpu 12.84s
52 Coqui XTTS-v2 cpu 13.26s
53 VibeVoice Realtime 0.5B cpu 14.99s
54 ZipVoice 123M cpu 15.03s
55 VibeVoice 7B cuda 15.16s
56 Chatterbox cpu 16.17s
57 OuteTTS 1.0 1B cuda 16.87s
58 Miso TTS 8B cuda 17.67s
59 Magpie-TTS cpu 33.05s
60 VoxCPM2 2B cpu 33.64s
61 Qwen3-TTS 1.7B Base cpu 36.51s
62 Fish Speech S2-Pro cuda 38.87s
63 Mars5-TTS cpu 50.17s
64 F5-TTS v1 cpu 50.71s
65 Zonos v0.1 cpu 50.77s
66 Mars5-TTS cuda 51.80s
67 VibeVoice 1.5B cpu 62.28s
68 Fish Speech 1.5 cpu 63.58s
69 Sesame CSM-1B cpu 70.00s

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 49ms
2 Piper cpu 109ms
3 NeuTTS Nano cuda 257ms
4 Pocket-TTS cpu 283ms
5 NeuTTS Nano cpu 338ms
6 OpenVoice v2 cuda 349ms
7 Kokoro cpu 668ms
8 OmniVoice cuda 683ms
9 Coqui XTTS-v2 cuda 923ms
10 Supertonic 3 cpu 993ms
11 Qwen3-TTS 1.7B (CUDA-graph) cuda 1.41s
12 Higgs Audio v3 TTS cuda 1.64s
13 VoxCPM2 2B cuda 1.82s
14 OpenVoice v2 cpu 1.96s
15 MOSS-TTS v1.5 cuda 2.90s
16 MOSS-TTS v1.0 cuda 3.05s
17 MOSS-TTS-Nano cuda 3.06s
18 Magpie-TTS cuda 3.32s
19 MOSS-TTS-Nano cpu 3.71s
20 Fish Speech 1.5 cuda 3.97s
21 Zonos v0.1 cuda 4.04s
22 Qwen3-TTS 1.7B Base cuda 4.80s
23 dots.tts (soar) cuda 5.67s
24 Coqui XTTS-v2 cpu 6.18s
25 OmniVoice cpu 11.73s
26 OuteTTS 1.0 1B cuda 12.05s
27 ZipVoice 123M cpu 12.88s
28 VoxCPM2 2B cpu 16.92s
29 Magpie-TTS cpu 27.26s
30 Qwen3-TTS 1.7B Base cpu 27.66s
31 Zonos v0.1 cpu 31.64s
32 Fish Speech 1.5 cpu 56.98s