Source code</>
UNREEL
Experimental
UNREEL
An open-source, self-hosted AI streaming service that writes scenes and renders MiniMax H3 Max Turbo video through fal as a channel plays.
Open source
Open-source AI live projects expose the code behind continuous generation, chat input, clip queues, and broadcast output. Most require your own keys and runtime.
A focused doorway into AI Live Hub’s guide to channels, platforms, creator tools, and open-source broadcast projects.
Selected signals
← All signalsProjects carrying this thread.
Source code</>
LAST FRAME / Interdimensional Game
Experimental
LAST FRAME / Interdimensional Game
An open-source playable-film prototype where branches are prepared ahead of a choice and the selected scene can appear nearly immediately.
Source code</>
infinite-livestream
Experimental
infinite-livestream
An end-to-end reference implementation where Twitch or YouTube prompts become expanded scenes, generated clips, and a continuous RTMP broadcast.
Source code</>
infinite-tv
Experimental
infinite-tv
An open-source chat-driven AI video system for Twitch or RTMP, using LTX Video locally or through fal, with a dashboard and metrics.
Source code</>
SlopTV
Experimental
SlopTV
An open-source loop that turns YouTube chat comments into video prompts, renders short clips with a local GPU workflow, and sends them back via RTMP.
Source code</>
mirrorstage
Experimental
mirrorstage
A modular self-hosted platform for turning chat or input into talking-head video, with configurable LLM, TTS, video, and OBS streaming providers.
Source code</>
infinite-slop-engine
Experimental
infinite-slop-engine
A one-day open-source experiment that moderates and rewrites YouTube chat before generating short H3 Max clips for a live broadcast.
Source code</>
Daydream Scope
Experimental
Daydream Scope
A workbench for running real-time generative video pipelines, with model selection, composable processing and WebRTC, NDI, Spout or Syphon output paths.
Source code</>
StreamDiffusion
Experimental
StreamDiffusion
A diffusion pipeline with browser demos that transform webcam or screen-capture frames as they arrive.
Source code</>
StreamDiffusionV2
Experimental
StreamDiffusionV2
An interactive video-diffusion system with released checkpoints, configurable inference and support for different GPU arrangements.
Source code</>
Live2Diff
Experimental
Live2Diff
A video-to-video research pipeline using one-way temporal attention to transform live input, with webcam and screen-video demos.
Source code</>
Self Forcing
Experimental
Self Forcing
Research code that trains video diffusion models on their own autoregressive rollouts and provides streaming-generation inference.
Source code</>
LongLive
Experimental
LongLive
A long-video research project whose 1.0 implementation supports sequential prompts and interactive generation, with a newer 2.0 infrastructure in the main repository.
Source code</>
Matrix-Game
Experimental
Matrix-Game
A family of interactive world models, with the 3.0 release focused on streaming generation and longer-horizon memory.
Source code</>
LiveTalking
Experimental
LiveTalking
A digital-human conversation engine supporting several lip-sync backends, interruption handling and WebRTC, RTMP or virtual-camera output.
Source code</>
Open-LLM-VTuber
Experimental
Open-LLM-VTuber
A voice-interactive AI companion with visual perception and a Live2D character, supporting local and hosted model backends.
Source code</>
MuseTalk
Experimental
MuseTalk
A lip-sync model that updates a prepared face video from incoming audio clips, with a documented real-time inference workflow.
Source code</>
FasterLivePortrait
Experimental
FasterLivePortrait
A community implementation of LivePortrait with ONNX and TensorRT execution plus a documented real-time camera workflow.
Source code</>
LivePortrait
Experimental
LivePortrait
A portrait-animation model that transfers movement from a driving performance into a source image or video.
Source code</>
Ditto TalkingHead
Experimental
Ditto TalkingHead
A talking-head synthesis project with online and offline streaming pipelines and released inference models.
Source code</>
DH_live
Experimental
DH_live
A lightweight digital-human project with browser inference examples and a real-time conversation server connecting speech recognition, an LLM and speech synthesis.
Source code</>
Ultralight Digital Human
Experimental
Ultralight Digital Human
A compact audio-driven digital-human model trained from footage of an individual, with mobile real-time inference described by its author.
Source code</>
OpenAvatarChat
Experimental
OpenAvatarChat
A modular avatar-chat system with replaceable speech recognition, language-model, speech synthesis and avatar components.
Source code</>
SoulX-FlashHead
Experimental
SoulX-FlashHead
A streaming talking-head research system with released Lite and Pro model variants and a Gradio streaming mode.
Source code</>
SoulX-LiveAct
Experimental
SoulX-LiveAct
A human-animation research system for streaming interactions, with memory compression and controls for action or emotion.
Source code</>
LiveKit Agents
Experimental
LiveKit Agents
An agent framework with WebRTC clients and examples connecting conversational services to external video-avatar providers.
Source code</>
Pipecat
Experimental
Pipecat
A Python framework for real-time conversational agents that orchestrates audio, video, AI services and streaming transports.
Source code</>
Vision Agents by Stream
Experimental
Vision Agents by Stream
A framework for building agents that watch and listen to live video, combining perception models with real-time conversation.
Source code</>
FastRTC
Experimental
FastRTC
A library for turning Python functions into real-time audio or video streams, with built-in UI and FastAPI integration.
Source code</>
TalkingHead
Experimental
TalkingHead
A JavaScript class for real-time lip-sync and full-body 3D avatars, with examples using AI speech and conversational backends.
Source code</>
NVIDIA Audio2Face-3D
Experimental
NVIDIA Audio2Face-3D
A collection of models, SDKs and integrations that turn recorded or streamed audio into lip-sync and expressive 3D facial animation.
Source code</>
LiteAvatar
Experimental
LiteAvatar
An audio-to-face model for 2D chat avatars, with sample avatar data and an interactive integration through OpenAvatarChat.
Source code</>
Oasis 500M
Experimental
Oasis 500M
The released smaller Oasis model and inference code generate gameplay frames conditioned on actions and an initial frame.
Source code</>
WorldCupVoice
Experimental
WorldCupVoice
An open-source project that watches an Agora RTC live feed, generates context-aware voice commentary, and publishes the AI audio back to viewers with a transcript.
AI Live Hub (5)AI live streaming (28)AI television (15)H3 Max (12)Audience-directed (11)Chat-driven (4)Interactive AI TV (6)Video generation (19)World building (8)Streamer tools (8)Real-time transforms (11)OBS (4)24/7 channels (5)Pre-generated TV (3)FAST channels (1)Roku (1)AI hosts (28)Live shopping (2)Ecommerce (2)AI news (2)YouTube (4)Self-host (34)Interactive video (8)Playable film (6)Twitch (2)RTMP (5)LTX Video (1)Talking heads (24)Moderation (1)Creator economy (3)AI advertising (3)World models (1)Multiplayer (2)Horror (1)