zai-org/RealVideo
A real-time streaming conversational video system that transforms text interactions into continuous, high-fidelity video responses using autoregressive diffusion.
WebSocket-based architecture integrating GLM-4.5-AirX for dialogue and GLM-TTS for speech synthesis, paired with autoregressive diffusion video generation for lip-synced avatar responses. Supports voice cloning from uploaded audio and runs on multi-GPU setups (minimum 2×80GB) with VAE and DiT services distributed across available compute, achieving sub-500ms frame generation for smooth real-time playback.
308 stars.
Stars
308
Forks
43
Language
Python
License
Apache-2.0
Category
Last pushed
Dec 15, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/zai-org/RealVideo"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
macrocosm-os/apex
SN1: An incentive mechanism for internet-scale conversational intelligence
jgravelle/pocketgroq
PocketGroq is a powerful Python library that simplifies integration with the Groq API, offering...
uezo/chatmemory
The simple yet powerful long-term memory manager between AI and you💕
Lin-jun-xiang/agent-line-bot
🤖Free Agent Line Bot with Google Image Search, Image Generator, Video Generator...
CORE-Labet/CORE
CORE is a plug-and-play conversational agent for any recommender system.