Inworld TTS-1-Max overview
Inworld TTS-1-Max by Inworld AI
Inworld TTS-1-Max is a high-fidelity, transformer-based neural text-to-speech model optimized for interactive and emotionally expressive voice synthesis.
Inworld TTS-1-Max specs
- Model ID
inworld/tts-1-max- Developer
- Inworld AI
- Input
- Text
- Output
- Audio
- Endpoint
POST https://api.aimlapi.com/v1/tts
Inworld TTS-1-Max API Overview
Inworld TTS-1-Max is a cutting-edge Transformer-based autoregressive text-to-speech (TTS) model developed to deliver unparalleled speech quality and expressiveness. With 8.8 billion parameters, it targets demanding professional and commercial applications requiring high-resolution, nuanced speech synthesis.
Technical Specifications
- Architecture: Transformer-based autoregressive model
- Parameters: 8.8 billion (largest in the TTS-1 family)
- Audio Output: High-resolution 48 kHz speech
- Supported Languages: 11 major languages
- Inference Speed: Approx. 8,000 tokens/sec per GPU on a 32 H100 setup
Performance Benchmarks
The TTS-1-Max model is currently ranked as a top performer on independent quality leaderboards.
Key Features
- Large-scale parameterization for superior voice naturalness and expressiveness
- Multilingual synthesis with high fidelity in diverse languages
- Emotional modulation capabilities enabling nuanced speech styles
- Non-verbal sounds and vocalization support enhance speech realism
- Pure reliance on in-context learning for voice cloning without pre-recorded speaker data
Comparison with Other Models
vs Inworld TTS-1: TTS-1-Max delivers superior expressiveness and naturalness thanks to its larger 8.8B parameter scale compared to TTS-1's 1.6B, ideal for premium content like audiobooks. However, TTS-1 prioritizes real-time speed at ~153 characters/second versus TTS-1-Max's ~69 characters/second, making it better for interactive apps.
vs ElevenLabs Multilingual V2: TTS-1-Max edges out with 59.1% head-to-head win rates in quality tests, offering finer emotional granularity and non-verbal sounds via markups. ElevenLabs provides strong multilingual cloning but lags in raw audio resolution and in-context learning purity.
vs MiniMax-Speech: TTS-1-Max prioritizes peak voice quality and 11-language fidelity over MiniMax's broader 32-language zero-shot cloning emphasis. While MiniMax shines in rapid one-shot replication, TTS-1-Max leads in benchmarked naturalness and emotional prosody control.
Use cases
- Professional voice-over and dubbing requiring premium quality
- Advanced conversational AI and virtual assistants with expressive speech
- Multilingual media content production and localization
- Interactive voice applications demanding real-time performance with rich emotional expression
- Audio books, gaming, and immersive virtual environments with diverse voice requirements
Inworld TTS-1-Max API pricing
| Usage | Price | Unit |
|---|---|---|
| Speech generation | $0.013 | per generation |
Prices from the AI/ML API price list, checked on Nov 4, 2026.
Inworld TTS-1-Max vs other speech models: prices
AI/ML API rates for 5 models, per generation.
| Model | Per generation |
|---|---|
 Inworld TTS-1-Max | $0.013 |
| Inworld TTS-1 | $0.007 |
 Qwen3 TTS Flash | $0.013 |
| Inworld TTS-1.5-Mini | $5.25 |
| Inworld TTS-1.5-Max | $10.5 |
How to use the Inworld TTS-1-Max API
View docs
Build with your agent
Give your agent the model details, code and documentation.
1Get your API key
Create API keyexport AIMLAPI_KEY="8f2b4c7d9a1e6b3f…"2Make your first request
curl 'https://api.aimlapi.com/v1/tts' \
-H "Authorization: Bearer $AIMLAPI_KEY" \
-H 'Content-Type: application/json' \
-d '{
"model": "inworld/tts-1-max",
"text": "Your order has shipped. It will arrive on Thursday between nine and eleven in the morning. Reply to this message if you need to change the delivery time."
}'import os
import requests
headers = {
"Authorization": f"Bearer {os.environ['AIMLAPI_KEY']}",
"Content-Type": "application/json",
}
payload = {
"model": "inworld/tts-1-max",
"text": "Your order has shipped. It will arrive on Thursday between nine and eleven in the morning. Reply to this message if you need to change the delivery time."
}
response = requests.post("https://api.aimlapi.com/v1/tts", headers=headers, json=payload, timeout=120)
response.raise_for_status()
print(response.json())const response = await fetch("https://api.aimlapi.com/v1/tts", {
method: "POST",
headers: {
Authorization: `Bearer ${process.env.AIMLAPI_KEY}`,
"Content-Type": "application/json",
},
body: JSON.stringify({
"model": "inworld/tts-1-max",
"text": "Your order has shipped. It will arrive on Thursday between nine and eleven in the morning. Reply to this message if you need to change the delivery time."
}),
});
if (!response.ok) throw new Error(`HTTP ${response.status}`);
console.log(await response.json());Models similar to Inworld TTS-1-Max
Related articles
-1%20(1).webp)
Inworld TTS-1-Max FAQ
What inputs and outputs does Inworld TTS-1-Max support?
Inworld TTS-1-Max takes text as input and returns audio.
How much does Inworld TTS-1-Max cost?
Inworld TTS-1-Max is priced at $0.013 / 1M tokens.
How is Inworld TTS-1-Max billed?
Inworld TTS-1-Max is billed per generation, a fixed charge per output rather than by prompt length.
Who created Inworld TTS-1-Max?
Inworld TTS-1-Max was built by Inworld AI.
How do I call Inworld TTS-1-Max on AI/ML API?
Send a request to https://api.aimlapi.com/v1/tts with inworld/tts-1-max as the model id.
Is Inworld TTS-1-Max available through a single API key?
Yes. Inworld TTS-1-Max is served through AI/ML API, so the same key and endpoint format used for other models applies.
What is the Inworld TTS-1-Max model ID?
Use inworld/tts-1-max as the model parameter. Requests go to https://api.aimlapi.com/v1/tts with your AI/ML API key.