Elite Agent Postcall Outputs
Emotion Tags
normalslowcryingsleepysighchuckle
Guided Emotion and Intonation
Input Streaming for Lower Latency
LLM-based Customizability
Model Tokenizer Type
Non-streaming (CNN-based) tokenizer
Native System Integration
Open Source Release Planned
Orpheus Speech Models
Medium (3B)Small (1B)Tiny (400M)Nano (150M)
Pretrained and Finetuned Models
Pretrained modelsFinetuned models
Rookie Agent Postcall Outputs
Sample Finetuning Scripts
Simultaneous Call Handling
100
Sliding Window Detokenizer
Streaming Inference Speed
Faster than playback on A100 40GB for 3B model
Supported Use Cases
Lead GenerationCustomer ServiceTechnical SupportReservationsConversational IVR
Training Data Volume
100k+ hours of speech, billions of text tokens
Unlimited Tools Per Agent
Vanguard Agent Postcall Outputs
Vanguard Agent Starter Pack