OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Chat & Assistants

Wraiter

Wraiter is a local Python application for collaborative fiction writing, in which a user and an AI take turns continuing a story. It supports typed input, microphone input, selectable AI-generated actions, character conversation, spoken narration and story illustrations.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Wraiter — OpenViST project profile in Chat & Assistants.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Wraiter when you want to build or explore fiction interactively on a local Python application, rather than submit a single prompt for a standalone text response.

Who it is for

  • Writers who want an AI to continue fiction between their own contributions.
  • People who want to steer a story through suggested actions or microphone input.
  • Users seeking a locally run story-writing application with spoken narration and optional illustrations.

Documented uses

  • Write a story by alternating user paragraphs with AI continuations.
  • Continue fiction by choosing among AI-generated actions or speaking through microphone input.
  • Chat with a fictional character and save, load, continue or revert a story.
  • Listen to narration with different voices for dialogue and story text.
  • Display an illustration related to a story.
Before you choose

Requirements & limits

The repository is public and identifies the project as open source under the MPL-2.0 license. Local setup requires Python 3.10 or later.

  • The README describes the writing model as replaceable with Hugging Face models through a command-line argument.
  • Hardware guidance lists at least 8 GB of GPU VRAM, or 28 GB of RAM for CPU text generation; GPU acceleration uses PyTorch with CUDA or ROCm.
  • The README says the application retains roughly the last thousand words, while the model input limit varies and is described as 2,048 words; older story events may drop out.
  • The project says it saves stories on the user's hard drive and connects to the internet to download models and install Python packages.

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 Wraiter locally

Prepare the local application using the repository's documented installation steps.

Before you start

  • Python 3.10 or later
  • A downloaded or cloned copy of the Wraiter repository
  • PowerShell on Windows or a shell for the corresponding installer script

Steps

  1. Download or clone the repository.

  2. Run install.ps1 in Windows PowerShell, or run install.sh in a shell.

  3. For GPU acceleration, set up PyTorch with CUDA or ROCm; alternatively, disable GPU in settings.json created after the first run.

Expected outcome

The documented setup is complete; the README identifies play.ps1 or play.sh as the next launch stage.

Guide limitations

  • The README states that GPU acceleration requires PyTorch configured with CUDA or ROCm.
  • The README gives 8 GB of GPU VRAM or 28 GB of RAM for CPU text generation as hardware guidance; actual performance is not guaranteed.
Documentation 1 ↗
Evidence and transparency

Sources & review

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