Official

minimax / video-01-director

Generate videos with specific camera movements

  • Public
  • 1.5K runs
Run with an API
Playground API Examples README

Examples

View more

Pricing

Official model
Pricing for official models works differently from other models. Instead of being billed by time, you’re billed by input and output, making pricing more predictable.
Learn more

This model is priced by how many videos are generated.

Check out our docs for more information about how per-video pricing works on Replicate.

Readme

T2V-01-Director

Overview

T2V-01-Director is an advanced AI video generation model that specializes in creating high-definition videos with cinematic camera movements. The model supports 720p resolution at 25fps and can generate videos up to 6 seconds in length.

Camera Movement Control System

Basic Usage

Camera movement instructions can be incorporated in two ways:

  1. Using bracketed commands: [Camera Movement]
  2. Natural language descriptions within the prompt

Examples

  • [Push in] A lamb stands alone in a snowy field, snowflakes gently falling around it.
  • [Pan left, Pan right] A woman in business attire sits at a café, sipping coffee while working on her laptop.
  • A man picks up a leather-bound book from a dusty shelf [Pedestal up], then begins reading it by candlelight [Static shot].
  • Above an ancient battlefield [Pan right, Pedestal down], a close-up reveals a warrior woman’s face, dirt-streaked but determined.
  • [Tracking shot] A red fox moves stealthily through autumn woods, hunting for prey.
  • [Push in, Zoom out] A surprised woman stands in a crowded beach scene, her expression frozen as the background comes into focus.
  • The shot begins from a dark rooftop [Pedestal up, Pan left], revealing a glittering cityscape with a large red moon hanging in the night sky.
  • [Pan left, Tilt up] Starting from a small rabbit looking upward, following its gaze to a bright red balloon floating into the sky.
  • [Pull out, Tilt up] A knight stands amid ruins holding a broken sword, the camera revealing burning city walls and a massive dragon shadow in the sky.
  • [Pan left, Pan right, Tracking shot] A woman in a flowing red dress walks confidently through a modern subway station.

Command Format

  • Single movement: [Movement]
  • Multiple simultaneous movements: [Movement1, Movement2]
  • Sequential movements: prompt [Movement1], second part of prompt [Movement2]

Supported Camera Movements

Horizontal Movements:

  • [Truck left] / [Truck right]: Camera moves horizontally
  • [Pan left] / [Pan right]: Camera rotates horizontally while stationary

Vertical Movements:

  • [Pedestal up] / [Pedestal down]: Camera moves vertically
  • [Tilt up] / [Tilt down]: Camera rotates vertically while stationary

Depth Movements:

  • [Push in] / [Pull out]: Camera moves forward/backward
  • [Zoom in] / [Zoom out]: Lens focal length changes

Special Movements:

  • [Tracking shot]: Camera follows subject
  • [Shake]: Camera vibration effect
  • [Static shot]: Fixed camera position

Best Practices

Movement Combinations:

  • Limit to 3 combined movements for optimal results
  • Place movement commands at the specific point in the prompt where the movement should occur
  • Use comma separation for simultaneous movements
  • Use sequential placement for sequential movements

Natural Language Integration:

  • Example: “The shot begins with a push in towards the subject [Push in], then pans across the landscape [Pan right] to reveal the sunset.”

Preset Shot Combinations

Circling Shots:

  • Left circling: [truck left, pan right, tracking shot]
  • Right circling: [truck right, pan left, tracking shot]

Walking Shots:

  • Left walking: [truck left, tracking shot]
  • Right walking: [truck right, tracking shot]

Dynamic Shots:

  • Upward tilt: [push in, pedestal up]
  • Scenic shot: [truck left, pedestal up]
  • Stage shots: [pan right, zoom in], [pan left, zoom in]
  • Downward tilt: [pedestal down, tilt up]

Technical Details

Model Capabilities

  • Video Resolution: 720p
  • Frame Rate: 25fps
  • Maximum Duration: 6 seconds
  • Input Types: Text-to-video and image-to-video

Usage Notes

For optimal results:

  • Be specific in camera movement descriptions
  • Place movement commands at logical points in the prompt
  • Combine natural language with bracketed commands for precise control
  • Keep combined movements to 3 or fewer

Prompt Structure:

  • Start with clear subject description
  • Include camera movements in brackets where they should occur
  • Add additional details about style, mood, or environment
  • Consider sequential timing of movements

Privacy policy

Data from this model is sent from Replicate to MiniMax.

Check their Privacy Policy for details:

https://intl.minimaxi.com/protocol/privacy-policy

Terms of Service

https://intl.minimaxi.com/protocol/terms-of-service