Fish Audio Drama 3
Speech and audio
drama-3
$2 per 1M UTF-8 bytes
Fish Audio S2 Pro / S2.1 Pro
Model ids2.1-pro
Price$15 per 1M UTF-8 bytes
Checked29 September 2026
Where it stands
64newer speech and audio models since its release
- NVIDIA Nemotron 3.5 Transcribe (30B-A4B)28 Oct 2026
- Hy-SimulMT-30B-A3B22 Oct 2026
- Eleven Realtime v1 Max / Mini20 Oct 2026
- Voxtral Mini Transcribe 3 (+ Mini Transcribe Realtime 3)20 Oct 2026
- Voxtral Small Transcribe 320 Oct 2026
The 5 newest are listed. All speech and audio models
- After it from Fish Audio
- Fish Audio Drama 313 Oct 2026
- Weights
- Weights are public. Licence: Fish Audio Research License. See the licence for commercial terms.
How to call it
6 ways to use Fish Audio S2 Pro / S2.1 Pro, with the model id each provider expects.
| Provider | Model id | Endpoint | Docs |
|---|---|---|---|
| Fish Audio API | s2.1-pro |
https://api.fish.audio/v1/tts (model passed in `model` header) |
docs.fish.audio |
| Fish Audio API (free tier) | s2.1-pro-free |
https://api.fish.audio/v1/tts |
|
| Fish Audio API | s2-pro |
||
| Hugging Face | fishaudio/s2-pro |
huggingface.co | |
| OpenRouter | openrouter.ai | ||
| GitHub | github.com |
Pricing
| Per 1M UTF-8 bytes | $15 |
|---|
As published: “USD per 1M UTF-8 bytes (s1, s2-pro, s2.1-pro); s2.1-pro-free $0 under fair use through 2026-11-30; ASR transcribe-1 $0.36/hr”.
Source: docs.fish.audio, checked 29 September 2026.
Specs
- Input
- Text and audio
- Output
- Audio
- Context window
- Not published
- Max output
- Not published
- Open weights
- Yes
- Licence
- Fish Audio Research License (S2 Pro weights; non-commercial, commercial license on request)
- Released
- 9 March 2026
- Status
- Current
- Type
- Speech and audio
What stands out
What its maker marketed at launch, and what people found later, each with its source.
Inline natural-language emotion/paralinguistic tags
Free-form bracket cues like [whisper], [laugh], [emphasis]; multi-speaker dialogue in one pass; 80+ languages from 10M+ hours of training audio.
Open model with production inference stack
Dual-AR (4B slow + 400M fast) on a Qwen3-4B backbone released with fine-tuning code and SGLang serving; RTF 0.195, ~100 ms TTFA; Seed-TTS Eval WER 0.54% zh / 0.99% en.
Free production API (S2.1 Pro)
Found after launch
S2.1 Pro (closed, 2026-06-23) offered free under fair use with ~90 ms TTFA, 83 languages; 61% win rate vs S2 Pro.
Notes
S2 Pro held #1 open-weights on Artificial Analysis until Breeze TTS 2 (Aug 2026); now #2 open (~1119 Elo). S2.1 Pro weights are NOT open. OpenRouter lists S2.1 Pro release as 2026-07-29 (API availability there). Predecessor OpenAudio S1 (s1) is deprecated and retires 2026-12-31 (requests then served and billed as s2.1-pro). Successor: Drama 3 (2026-10-13, see fish-audio-drama-3); on Artificial Analysis S2 Pro is now the #3 open-weights model (Elo 1118, read 2026-10-13). The pricing page no longer shows an end date for s2.1-pro-free (read 2026-10-13).
Sources: https://docs.fish.audio/developer-guide/models-pricing/pricing-and-rate-limits , https://huggingface.co/fishaudio/s2-pro , https://fish.audio/blog/s2-1-pro-free-api/