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

TTS Bench — Samples — linux-cloning

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: cloning · chris_hemsworth voice (note: metavoice uses the 67s clip — it needs ≥30s ref — all other models use the 15s clip)
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 Pocket-TTS cpu 81ms
2 StyleTTS 2 cpu 142ms
3 StyleTTS 2 cuda 145ms
4 LuxTTS cuda 195ms
5 OpenVoice v2 cuda 254ms
6 NeuTTS Nano cuda 359ms
7 NeuTTS Nano cpu 418ms
8 NeuTTS Air cuda 432ms
9 MiraTTS cuda 453ms
10 Coqui XTTS-v2 cuda 475ms
11 NeuTTS Air cpu 559ms
12 Chatterbox Turbo cuda 631ms
13 Qwen3-TTS 1.7B (CUDA-graph) cuda 767ms
14 Higgs Audio v3 TTS cuda 856ms
15 OpenVoice v2 cpu 988ms
16 longcat_1b cuda 1.12s
17 Chatterbox cuda 1.13s
18 OmniVoice cuda 1.14s
19 F5-TTS v1 cuda 1.14s
20 VoxCPM2 2B cuda 1.22s
21 LuxTTS cpu 1.29s
22 Dia 1.6B-0626 cuda 1.29s
23 Fish Speech 1.5 cuda 1.93s
24 MOSS-TTS-Nano cuda 1.97s
25 MOSS-TTS v1.5 cuda 2.19s
26 MOSS-TTS v1.0 cuda 2.25s
27 Zonos v0.1 cuda 2.34s
28 MetaVoice-1B cuda 2.51s
29 IndexTTS-2 cpu 2.80s
30 longcat_3p5b cuda 3.03s
31 IndexTTS-2 cuda 3.09s
32 Step-Audio-EditX cuda 3.10s
33 Coqui XTTS-v2 cpu 3.15s
34 Sesame CSM-1B cuda 3.34s
35 dots.tts (soar) cuda 3.49s
36 MOSS-TTS-Nano cpu 3.52s
37 Chatterbox Turbo cpu 3.91s
38 VibeVoice 7B cuda 4.00s
39 Echo-TTS cuda 4.34s
40 OuteTTS 1.0 1B cuda 4.67s
41 Miso TTS 8B cuda 4.86s
42 ZipVoice 123M cpu 6.97s
43 Chatterbox cpu 8.01s
44 DramaBox cuda 9.77s
45 Fish Speech S2-Pro cuda 10.52s
46 VoxCPM2 2B cpu 14.19s
47 Zonos v0.1 cpu 18.48s
48 Fish Speech 1.5 cpu 19.19s
49 Sesame CSM-1B cpu 27.72s
50 OmniVoice cpu 35.63s
51 F5-TTS v1 cpu 45.27s
52 Mars5-TTS cpu 68.86s
53 Mars5-TTS cuda 70.40s

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 Pocket-TTS cpu 110ms
2 StyleTTS 2 cuda 197ms
3 StyleTTS 2 cpu 205ms
4 LuxTTS cuda 219ms
5 OpenVoice v2 cuda 272ms
6 NeuTTS Nano cuda 343ms
7 NeuTTS Nano cpu 378ms
8 NeuTTS Air cuda 426ms
9 NeuTTS Air cpu 598ms
10 MiraTTS cuda 814ms
11 Coqui XTTS-v2 cuda 998ms
12 Chatterbox Turbo cuda 1.17s
13 longcat_1b cuda 1.20s
14 OmniVoice cuda 1.35s
15 F5-TTS v1 cuda 1.50s
16 Higgs Audio v3 TTS cuda 1.76s
17 LuxTTS cpu 2.08s
18 Chatterbox cuda 2.21s
19 OpenVoice v2 cpu 2.22s
20 MOSS-TTS-Nano cuda 2.27s
21 VoxCPM2 2B cuda 2.38s
22 MOSS-TTS v1.0 cuda 3.14s
23 MOSS-TTS v1.5 cuda 3.19s
24 longcat_3p5b cuda 3.24s
25 IndexTTS-2 cpu 4.25s
26 Echo-TTS cuda 4.32s
27 IndexTTS-2 cuda 4.62s
28 Step-Audio-EditX cuda 4.62s
29 MetaVoice-1B cuda 5.34s
30 MOSS-TTS-Nano cpu 5.40s
31 Fish Speech 1.5 cuda 5.62s
32 Zonos v0.1 cuda 5.65s
33 Sesame CSM-1B cuda 6.49s
34 Coqui XTTS-v2 cpu 6.81s
35 Chatterbox Turbo cpu 7.40s
36 Dia 1.6B-0626 cuda 7.86s
37 VibeVoice 7B cuda 7.98s
38 dots.tts (soar) cuda 8.43s
39 DramaBox cuda 10.81s
40 OuteTTS 1.0 1B cuda 11.19s
41 ZipVoice 123M cpu 13.29s
42 Chatterbox cpu 14.36s
43 Miso TTS 8B cuda 19.99s
44 VoxCPM2 2B cpu 26.35s
45 Fish Speech S2-Pro cuda 27.16s
46 Sesame CSM-1B cpu 38.09s
47 OmniVoice cpu 44.05s
48 Qwen3-TTS 1.7B (CUDA-graph) cuda 45.49s
49 Zonos v0.1 cpu 45.50s
50 F5-TTS v1 cpu 57.96s
51 Fish Speech 1.5 cpu 61.64s
52 Mars5-TTS cpu 72.55s
53 Mars5-TTS cuda 80.29s

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 Pocket-TTS cpu 135ms
2 LuxTTS cuda 332ms
3 NeuTTS Nano cuda 373ms
4 StyleTTS 2 cpu 412ms
5 StyleTTS 2 cuda 415ms
6 NeuTTS Nano cpu 449ms
7 NeuTTS Air cuda 531ms
8 OpenVoice v2 cuda 549ms
9 NeuTTS Air cpu 594ms
10 longcat_1b cuda 1.63s
11 OmniVoice cuda 2.23s
12 MiraTTS cuda 2.32s
13 F5-TTS v1 cuda 3.14s
14 Chatterbox Turbo cuda 3.32s
15 Coqui XTTS-v2 cuda 3.65s
16 Qwen3-TTS 1.7B (CUDA-graph) cuda 3.85s
17 Echo-TTS cuda 4.34s
18 Higgs Audio v3 TTS cuda 4.77s
19 longcat_3p5b cuda 5.50s
20 LuxTTS cpu 5.76s
21 Chatterbox cuda 6.05s
22 MOSS-TTS-Nano cuda 6.81s
23 MOSS-TTS v1.5 cuda 7.32s
24 VoxCPM2 2B cuda 7.42s
25 OpenVoice v2 cpu 7.54s
26 MOSS-TTS v1.0 cuda 7.68s
27 Step-Audio-EditX cuda 9.99s
28 IndexTTS-2 cpu 11.39s
29 IndexTTS-2 cuda 12.24s
30 Sesame CSM-1B cuda 13.52s
31 MOSS-TTS-Nano cpu 14.54s
32 MetaVoice-1B cuda 14.56s
33 DramaBox cuda 14.62s
34 Fish Speech 1.5 cuda 15.11s
35 Zonos v0.1 cuda 17.67s
36 Chatterbox Turbo cpu 21.26s
37 Dia 1.6B-0626 cuda 21.73s
38 VibeVoice 7B cuda 24.87s
39 Coqui XTTS-v2 cpu 25.42s
40 OuteTTS 1.0 1B cuda 30.19s
41 dots.tts (soar) cuda 36.35s
42 Chatterbox cpu 42.09s
43 VoxCPM2 2B cpu 67.86s
44 OmniVoice cpu 69.74s
45 Sesame CSM-1B cpu 80.94s
46 Fish Speech S2-Pro cuda 81.01s
47 Miso TTS 8B cuda 84.98s
48 Mars5-TTS cuda 104.33s
49 Mars5-TTS cpu 111.83s
50 F5-TTS v1 cpu 141.44s
51 Fish Speech 1.5 cpu 143.99s
52 Zonos v0.1 cpu 164.17s

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 Pocket-TTS cpu 108ms
2 LuxTTS cuda 220ms
3 StyleTTS 2 cuda 254ms
4 StyleTTS 2 cpu 264ms
5 NeuTTS Nano cuda 366ms
6 OpenVoice v2 cuda 384ms
7 NeuTTS Nano cpu 428ms
8 NeuTTS Air cuda 523ms
9 NeuTTS Air cpu 597ms
10 Chatterbox Turbo cuda 1.32s
11 MiraTTS cuda 1.35s
12 OmniVoice cuda 1.40s
13 longcat_1b cuda 1.41s
14 F5-TTS v1 cuda 1.56s
15 Coqui XTTS-v2 cuda 1.70s
16 LuxTTS cpu 2.24s
17 Higgs Audio v3 TTS cuda 2.52s
18 OpenVoice v2 cpu 2.63s
19 MOSS-TTS-Nano cuda 2.99s
20 Chatterbox cuda 3.04s
21 VoxCPM2 2B cuda 3.10s
22 MOSS-TTS v1.0 cuda 3.43s
23 longcat_3p5b cuda 3.91s
24 MOSS-TTS v1.5 cuda 3.92s
25 Echo-TTS cuda 4.34s
26 IndexTTS-2 cpu 5.00s
27 Fish Speech 1.5 cuda 5.13s
28 Step-Audio-EditX cuda 5.58s
29 MOSS-TTS-Nano cpu 5.89s
30 IndexTTS-2 cuda 6.00s
31 Zonos v0.1 cuda 6.11s
32 MetaVoice-1B cuda 7.53s
33 Chatterbox Turbo cpu 8.85s
34 Dia 1.6B-0626 cuda 8.86s
35 Coqui XTTS-v2 cpu 8.87s
36 Sesame CSM-1B cuda 9.56s
37 DramaBox cuda 10.95s
38 VibeVoice 7B cuda 11.79s
39 OuteTTS 1.0 1B cuda 12.30s
40 dots.tts (soar) cuda 12.70s
41 ZipVoice 123M cpu 15.77s
42 Miso TTS 8B cuda 17.72s
43 Chatterbox cpu 18.60s
44 VoxCPM2 2B cpu 37.91s
45 OmniVoice cpu 45.42s
46 Qwen3-TTS 1.7B (CUDA-graph) cuda 45.62s
47 Zonos v0.1 cpu 50.77s
48 Fish Speech S2-Pro cuda 57.62s
49 Sesame CSM-1B cpu 58.88s
50 F5-TTS v1 cpu 59.69s
51 Fish Speech 1.5 cpu 74.24s
52 Mars5-TTS cpu 90.10s
53 Mars5-TTS cuda 96.21s

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 Pocket-TTS cpu 320ms
2 NeuTTS Nano cuda 343ms
3 OpenVoice v2 cuda 352ms
4 NeuTTS Nano cpu 424ms
5 Coqui XTTS-v2 cuda 808ms
6 OmniVoice cuda 1.33s
7 VoxCPM2 2B cuda 1.37s
8 Qwen3-TTS 1.7B (CUDA-graph) cuda 1.41s
9 Higgs Audio v3 TTS cuda 1.42s
10 MOSS-TTS-Nano cuda 2.06s
11 OpenVoice v2 cpu 2.20s
12 MOSS-TTS v1.5 cuda 2.76s
13 MOSS-TTS v1.0 cuda 3.20s
14 dots.tts (soar) cuda 3.24s
15 Fish Speech 1.5 cuda 3.74s
16 MOSS-TTS-Nano cpu 3.79s
17 Zonos v0.1 cuda 3.97s
18 Coqui XTTS-v2 cpu 5.99s
19 OuteTTS 1.0 1B cuda 8.54s
20 ZipVoice 123M cpu 12.85s
21 VoxCPM2 2B cpu 18.72s
22 Zonos v0.1 cpu 31.28s
23 OmniVoice cpu 42.37s
24 Fish Speech 1.5 cpu 61.05s