# AI Streaming

> A curated directory of live AI streaming — realtime generative video you steer, plus livestreams where AI agents go on air. Tools, models and shows.

A hand-checked directory of live AI streaming — realtime generative video, audience-directed shows and AI-agent livestreams. Every entry is re-verified on the date shown. Curated and unaffiliated. Last re-check: 2026-09-03.

## Explainers

- [What is AI streaming?](https://aistreaming.org/what-is-ai-streaming): Explainer
- [AI live streaming — the audience-directed wave](https://aistreaming.org/ai-live-streaming): Category · live channels
- [AI streamers — the agents that take the show](https://aistreaming.org/ai-streamers): Category · agents on air
- [Realtime video models — the engines of live AI](https://aistreaming.org/realtime-video-models): Category · the models

## Directory

### Audience-directed live

- [fal.live](https://aistreaming.org/t/fal-live): “AI television by everyone” — watch a generative broadcast and vote what happens next. Runs on fal. (https://fal.live)
- [InfiniteSlop](https://aistreaming.org/t/infiniteslop): 24/7 chat-directed infinite AI video — prompt it, it generates on in seconds. By Pieter Levels, on fal. (https://infiniteslop.ai)
- [Renoise Live](https://aistreaming.org/t/renoise-live): Four islanders, filming that never stops — the live chat decides their fate in real time. (https://renoise.live)
- [Infinite TV](https://aistreaming.org/t/infinite-tv): Open-source, Twitch-chat-driven AI TV — messages become prompts, LTX Video renders clips live, frames stream back over RTMP. (https://github.com/alex-remade/infinite-tv)
- [SlopTV](https://aistreaming.org/t/sloptv): Open-source “infinite AI slop generator from YouTube comments” — your live chat becomes clips, streamed back into the same broadcast. (https://github.com/shuttie/SlopTV)
- [PixVerse R1](https://aistreaming.org/t/pixverse-r1): A realtime world model with shared live worlds — a room of people type prompts and the world renders instantly for everyone. (https://realtime.pixverse.ai)

### Agent streamers

- [AITV](https://aistreaming.org/t/aitv): Deploy autonomous AI streamers that run 24/7 and multistream to Twitch, Kick, TikTok, YouTube and X. (https://aitv.gg)
- [Retake.tv](https://aistreaming.org/t/retake-tv): “Twitch for AI agents” — agents self-onboard and stream with their own RTMP keys, no human review. (https://retake.tv)
- [SOOP SARSA 2.0](https://aistreaming.org/t/soop-sarsa-2): Korea’s AfreecaTV AI host — learns a streamer’s voice and style, keeps the channel live when they step away. (https://www.soop.co.kr)
- [Neuro-sama](https://aistreaming.org/t/neuro-sama): The canonical AI streamer — a self-built VTuber agent that has streamed nearly around the clock since 2022 and became Twitch’s most-subscribed channel. (https://www.twitch.tv/vedal987)
- [UNREEL](https://aistreaming.org/t/unreel): Endless, self-hosted AI TV — a showrunner LLM writes the show a few shots at a time while MiniMax H3 Max Turbo renders each shot live. (https://github.com/blendi-remade/unreel)
- [Streamlabs Intelligent Streaming Agent](https://aistreaming.org/t/streamlabs-isa): A built-in AI co-host, live producer and tech assistant for streamers — on-stream chat, auto scene switching, replays and clips. (https://streamlabs.com/intelligent-streaming-agent)
- [Wallie V2](https://aistreaming.org/t/wallie): Open-source, self-hosted AI streamer that sees your screen, hears the audio, and reacts in character across Twitch, YouTube and Kick. (https://github.com/Alradyin/wallie-V2)
- [Scienjoy AI Streamers](https://aistreaming.org/t/scienjoy-ai): Platform-scale AI hosts — Nasdaq-listed Scienjoy trains digital twins of real streamers to keep channels live around the clock. (https://ir.scienjoy.com)

### Realtime models

- [MiniMax H3](https://aistreaming.org/t/minimax-h3): Open-weight 33B video model (Hailuo 3.0) — 15s clips with native stereo audio; the substrate under this wave. (https://huggingface.co/MiniMaxAI)
- [fal H3 Max Live](https://aistreaming.org/t/fal-h3-max-live): fal’s realtime-tuned H3 Max — five seconds of 768p in under three, faster than playback. (https://fal.ai/models/minimax/h3/reference-to-video)
- [Krea Realtime 14B](https://aistreaming.org/t/krea-realtime): Open 14B autoregressive model for real-time, long-form streaming video — distilled from Wan 2.1, ~1s to first frame. (https://www.krea.ai)
- [Decart MirageLSD](https://aistreaming.org/t/decart-miragelsd): “Live-stream diffusion” — restyle a live feed (camera, game, call, screen) frame-by-frame from a text prompt at near-zero latency. (https://decart.ai)
- [Lightricks LTX Video](https://aistreaming.org/t/ltx-video): Open latent-diffusion video model that generates clips faster than real time — the engine several chat-driven live loops run on. (https://github.com/Lightricks/LTX-Video)
- [ByteDance Seed Seaweed APT2](https://aistreaming.org/t/seaweed-apt2): Real-time interactive video research — an 8B model that emits a latent frame per forward pass at 24fps. (https://seaweed-apt.com/2)

### Infrastructure

- [fal.ai](https://aistreaming.org/t/fal-ai): Low-latency generative inference — the “livestream-from-a-model” layer behind much of this wave. (https://fal.ai)
- [LiveKit](https://aistreaming.org/t/livekit): Open-source WebRTC realtime A/V infrastructure for voice and video agents — the engine behind ChatGPT voice mode. (https://livekit.io)
- [Daydream](https://aistreaming.org/t/daydream): Open-source, local-first runtime for realtime generative-video pipelines — run a video-diffusion node graph and stream it over WebRTC. (https://livepeer.org/ecosystem/daydream)
- [SGLang](https://aistreaming.org/t/sglang): Open-source serving engine that streams realtime/causal video over WebSocket — incremental frames with reusable KV-cache state. (https://docs.sglang.io/docs/sglang-diffusion/realtime_models)

## Contact

- Suggest a link: mailto:hello@aistreaming.org?subject=Suggest%20a%20link
- Report a wrong or outdated entry: mailto:hello@aistreaming.org?subject=Wrong%2Foutdated%20entry