OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Image & Video AI

Story2Board

Story2Board is an open-source, training-free Python framework that generates expressive storyboard panels from natural-language descriptions, using reference-latent anchoring and attention-value mixing to support subject continuity across changing scenes.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Story2Board — OpenViST project profile in Image & Video AI.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when you want to create a multi-panel image sequence from text and need a recurring subject across changing scenes.

Who it is for

  • Visual storytellers and illustrators creating text-driven storyboard previews.
  • Researchers exploring diffusion-based image generation for multi-panel narrative sequences.

Documented uses

  • Turn a subject and a series of scene descriptions into a storyboard.
  • Reuse a reference panel to help maintain character identity across other panels.
  • Explore varied compositions, settings and actions within one visual sequence.
Before you choose

Requirements & limits

The project is a public GitHub repository licensed under MIT. Its README lists Linux, Python 3.12 and CUDA 12.x as officially supported.

  • The documented workflow is command-line based and requires subject, reference-panel, additional-panel and output-directory arguments.
  • The project is training-free and describes continuity mechanisms that do not require architectural changes or fine-tuning.
  • Linux with Python 3.12 and CUDA 12.x is the documented supported setup; Windows and macOS are not officially tested.
  • The implementation acknowledges PyTorch, Hugging Face Diffusers and FLUX.1-dev weights.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install Story2Board on Linux

Prepare the documented Python environment and install the repository dependencies.

Before you start

  • Linux, Python 3.12 and CUDA 12.x are listed as officially supported.
  • Conda and the Git command-line tool.

Steps

  1. Clone the Story2Board repository and change into its directory.

  2. Create and activate a Conda environment using Python 3.12.

  3. Install dependencies with pip install -r requirements.txt.

Expected outcome

A Python environment with the repository dependencies installed.

Guide limitations

  • The documented installation steps do not themselves generate a storyboard.
  • Windows and macOS are not officially tested.
Documentation 1 ↗
Source-based practical guide

Generate a storyboard from panel prompts

Pass a subject, reference-panel description and additional panel prompts to the documented command-line entry point.

Before you start

  • A prepared Story2Board environment.
  • A subject description, a reference-panel prompt, one or more additional panel prompts and an output directory.

Steps

  1. Run main.py with --subject, --ref_panel_prompt, --panel_prompts and --output_dir.

  2. Use the reference prompt for the top panel and supply one or more prompts for the remaining panels.

Expected outcome

The documented workflow writes generated images to the output directory and logs constructed per-panel prompts; the reference panel anchors the recurring subject while other panels vary scenes or actions.

Guide limitations

  • The repository specifies Linux, Python 3.12 and CUDA 12.x as officially supported.
  • Windows and macOS are not officially tested.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
17 October 2026
Original website title
DavidDinkevich/Story2Board · Official homepage title · retained in its source language
How we review sourcesSuggest a correction