---
schema: 1
type: term
title: >-
  **Autoregressive video model** — a model that generates video frame by frame,
  feeding its previously generated frames back in to predict each next one, like
  an LLM's next-token prediction applied to video
description: >-
  **Autoregressive video model** — a model that generates video frame by frame,
  feeding its previously generated frames back in to predict each next one, like
  an LLM's next-token prediction applied to video.
canonical: >-
  https://liquidgold.cc/channel/everyinc/explore/term/autoregressive-video-model-a-model-that-generates-v-4bfb07ea0c97
markdown: >-
  https://liquidgold.cc/channel/everyinc/explore/term/autoregressive-video-model-a-model-that-generates-v-4bfb07ea0c97.md
llms: null
parent: 'https://liquidgold.cc/channel/everyinc/llms.txt'
channel: everyinc
updated: '2026-10-23T22:12:35.346Z'
language: null
children: []
---


# \*\*Autoregressive video model\*\* — a model that generates video frame by frame, feeding its previously generated frames back in to predict each next one, like an LLM's next-token prediction applied to video

Canonical: https://liquidgold.cc/channel/everyinc/explore/term/autoregressive-video-model-a-model-that-generates-v-4bfb07ea0c97

Parent: [Channel index](https://liquidgold.cc/channel/everyinc/llms.txt)

## Definition

**Autoregressive video model** — a model that generates video frame by frame, feeding its previously generated frames back in to predict each next one, like an LLM's next-token prediction applied to video.

In this channel's material, an **autoregressive video model** appears in the exploration of architectures beyond the standard text-prediction LLM — specifically in the discussion of **Mirage**, the real-time video-to-video system that generates a playable game world in about **40 milliseconds** per frame [^08615b13-5661-5b2b-bcb0-3e4b551b9b63@504].The analogy the channel draws is direct: just as a large language model predicts the *next token* by feeding everything it has already produced back in, an autoregressive video model predicts the *next frame* the same way — each newly generated frame becomes part of the input for generating the next one [^08615b13-5661-5b2b-bcb0-3e4b551b9b63@504]. That frame-by-frame feedback loop is what makes Mirage's world generation feel continuous rather than pre-rendered.Within the channel's treatment of model architectures, the autoregressive approach is one option among several — it changes _how_ the model produces output, and that change determines whether it fits the task. Real-time interactive worlds like Mirage lean on autoregressive frame prediction because latency matters frame to frame, whereas other architectures (such as non-autoregressive energy-based models) trade off differently [^08615b13-5661-5b2b-bcb0-3e4b551b9b63@504].- An **autoregressive video model** generates video one frame at a time, using previously generated frames as input for the next prediction.- The channel frames it as an **LLM next-token prediction applied to video**.- It is central to understanding **Mirage**, the real-time video-to-video game-world engine covered in this material [^08615b13-5661-5b2b-bcb0-3e4b551b9b63@504].