Drama 3
- Family
- Drama
- Released
- 2026-09-23
- Parameters
- 35B (3B activated)
- Architecture
- Mixture of Experts
- Specialization
- text-to-speech
- Openness
- Open weights
- License
- Fish Audio Community LicenseRestricted
- Weights
- Available
- Code
- Unknown
Fish Audio (Hanabi AI Inc.) builds expressive multilingual text-to-speech, voice cloning and speech-to-text models and s
WebsiteNo tracked provider pricing is available yet.
About
Drama 3 is Fish Audio's text-to-speech model for directed voice performance. Plain-language directions in square brackets before a line set emotion, intent and pacing, paired tags apply effects to specific words, and it handles sounds, pauses, multi-speaker dialogue and voice cloning across 83 languages. It is a 35B mixture-of-experts model with 3B active parameters per token. Fish Audio announced it as a preview on its API on September 23, 2026 and released the weights on October 13, 2026 under the Fish Audio Community License, which allows research and limited commercial use.
Providers(2)
Compare all 2Compare API pricing across 2 tracked provider routes for input and output tokens, batch, and cached reads when available.
| Provider | Input / 1M | Output / 1M | Route |
|---|---|---|---|
| OpenRouter | - | - | ServerlessPartial |
| Vercel AI Gateway | - | - | ServerlessPartial |
Capabilities
Benchmark peer barsfor Coding
No task-mapped benchmark peers are available for this model yet.
Migration checks
No linked migration route is available for this model yet.
API versions
Drama 3 Previewdrama-3drama-3-previewfishaudio/drama-3fish-audio/drama-3Fish Audio (Hanabi AI Inc.) builds expressive multilingual text-to-speech, voice cloning and speech-to-text models and s
WebsiteNo tracked provider pricing is available yet.