Back to News
NineNineSix•July 7, 2026
Gepard 1.0
Open Source
text-to-speech
voice-cloning
autoregressive
vllm
### TL;DR
Gepard 1.0 is an autoregressive, prosody-aware text-to-speech model designed specifically for real-time conversational AI. It utilizes a Qwen3.5-based backbone to generate audio in single-pass frames, enabling ultra-low latency and high-throughput streaming suitable for voice agents.
Key Insights & Metrics
Pricing
Free
Cost structure
Version
1.0
Current release version
Hardware
High-performance GPU recommended for real-time throughput (e.g., RTX 5090 for ~25x real-time; RTX Pro 6000 Blackwell for massive parallel scaling).
Compute requirements
Category
Open Source
Licensing model
Region
United States
Primary region
Key Features
- Single-pass, streaming-first frame generation for low-latency dialogue
- vLLM-native support allowing for high-throughput concurrent sessions
- Zero-shot voice cloning integrated into the prefill stage
- Classifier-free guidance (CFG) distilled into weights for quality at no extra compute cost
- Optimized for real-time voice agents with ~50ms time-to-first-audio
Discussion
0
Upvotes
0
Downvotes
0 reviews
Sign in to leave a review
Reviews
No reviews yet. Be the first to review!