whatmodel.sh / tts / F5-TTS v1 Base
F5-TTS v1 Base
↗ share
+ contribute
F5-TTS v1 Base is an open-weights text-to-speech model from SWivid (SJTU X-LANCE), released March 2025. Its published size is 0.337B parameters. Licence: cc-by-nc-4.0.
Catalogue identity notes
SWivid/F5-TTS F5TTS_v1_Base/model_1250000.safetensors (committed 2025-03-12). Distinct from the original F5TTS_Base (v0, 1.2M updates).
Model information sources
OUR COUNT from the safetensors header of the official F5TTS_v1_Base/model_1250000.safetensors (pinned revision): 337,096,836 F32 weights plus two scalar buffers. The Vocos vocoder is not included. The F5-TTS paper (arXiv 2410.06885v3) gives 335.8M for the (v0) Base model. Observed 2026-10-07.
Score within the Vapi Humanness · Bradley–Terry rating, 2026-09-25 snapshot comparison set only.
$ how it compares
The text-to-speech models ranked either side of F5-TTS v1 Base, out of 148 with a score, then more from SWivid (SJTU X-LANCE).
| # | model | score | where | ||
|---|---|---|---|---|---|
| 38 |  Eleven Flash v2.5 | 63±2 | api $50.00/M chars | ||
| 39 |  IndexTTS (v1) | 62±3 | local | ||
| 40 | F5-TTS v1 Base | 62±3 | local | ||
| 41 |  Inworld TTS-1.5 Max | 62±3 | api $10.00/M chars | ||
| 42 |  ZipVoice | 62±3 | local | ||
| more from SWivid (SJTU X-LANCE) | |||||
| 50 | E2 TTS Base (F5-TTS reproduction) | 60±2 | local | ||
| 60 | F5-TTS Base (v0) | 57±2 | local | ||
see all text-to-speech models · find the ones that fit your machine
$ what it's good at
voice cloning▁·▁▁▁▁▁▁▇█▃▃ #14 of 37 · 47% of its score · 4 tests
Beats Gemini 3.1 Flash TTS Preview, Fun-CosyVoice3-0.5B-2512, IndexTTS-2 here, though they rank higher overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| Seed-TTS test-en · human speaker-similarity MOS (SMOS) | 3.08 / 5 ✎ | ▄········█·▄ beats 2/3 | 17% |
| Seed-TTS test-en · speaker similarity (WavLM-large SV) | 68.0% ✎ | ▁·▂▂▂█▆▃▅··▂ beats 10/30 | 15% |
| Seed-TTS test-en · UTMOS (predicted MOS) | 3.66 / 5 ✎ | ▄█▄··▄···▄·▄ beats 5/6 | 7.6% |
| Seed-TTS test-en · Whisper large-v3 WER | 1.4% ✎ | ▁···▂▂▂▂█▆▄▂ beats 30/35 | 7.1% |
other tests5 not in a task · 50% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| LibriSpeech-PC test-clean (F5-TTS subset) · speaker similarity (WavLM SV) | 68.0% ✎ | ▄······▄·█▄▄ beats 4/5 | 18% |
| LibriSpeech-PC test-clean (F5-TTS subset) · UTMOS (predicted MOS) | 3.83 / 5 ✎ | █···█······█ beats 1/2 | 11% |
| LibriSpeech-PC test-clean (F5-TTS 4-10 s subset) · Whisper large-v3 WER | 2.0% ✎ | █····████··█ beats 5/5 | 11% |
| RVCBench LibriTTS clean voice cloning · ECAPA speaker similarity | 56.0% ✎ | ▃·····▂▂▂▃█▆ beats 8/15 | 10% |
| RVCBench LibriTTS clean voice cloning · Whisper-medium WER | 12.0% ✎ | ▂▂▂··▃·▃▃▃█▂ beats 3/15 | 0.1% |
$ where to run it
Other ways to run it
- Run it as published · needs ~1.35 GB · from github.com/SWivid (official)
pip install f5-tts && f5-tts_infer-cli --model F5TTS_v1_Base --gen_text "Hello, world."
Official weights and documentation: huggingface.co/SWivid ↗
$ how the score was worked out · DiSTAR: Diffusion over a Scal… 87%, RVCBench: Benchmarking Robust… 10%
Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works
| source | results | share of the score |
|---|---|---|
 DiSTAR: Diffusion over a Scalable Token Autoregressive Representation for Speech Generation (arXiv:2510.12210v2) | 7 | 87% |
| RVCBench: Benchmarking Robustness of Voice Cloning Across Modern Audio Generation Models (arXiv:2602.00443v2) | 2 | 10% |
| prior · pull toward the middle | 3% | |
| weighted average | = 62.1 |
$ improve this page
Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.
data updated · sources and licences · json