Building Real-time AI Generated Worlds in TouchDesigner
This course has limited capacity, reserve your spot now to secure your place!
In this workshop, participants will learn to use real-time image and text generation technology to construct a virtual environment and generative narrative. We will look at some examples of AI-generated worlds, AI worldbuilding tools, and creative and critical approaches to working with AI tools to inspire your work. Then, we’ll dive into working with a large language model (LLM) and prompt engineering to drive the narrative generation for students to create their own virtual worlds. Working with TouchDesigner and some simple Python scripting, students will then use real-time image generation to render a visual world that evolves together with their evolving narrative.
Course Logistics
Dates: Saturday, June 21, 2025
Times: 1 – 4PM PT
Cost: $120
Format:
In-Person Workshop at Gray Area
Address:
2665 Mission St, San Francisco
Experience Level: Intermediate to Advanced
Prerequisites:
Some familiarity with TouchDesigner and Python scripting would be helpful. Coding experience would be helpful.
Requirements:
• Bring your laptop and charger
• Set up a fal.ai API key beforehand (you will need a github account). I recommend buying $5-10 credits. (Minimum $5)
• Set up OpenAI developer account and an API key. (Starting credits should be enough).
• Install TouchDesigner 099, 2023
• Install Python 3.11 on your machine
Additional Information:
• No Refunds or Exchanges.
• View our FAQ here.
• Contact [email protected] with any questions.
Workshop Outline
- Introduction
- Looking at relevant tools and existing works
- Narrative and Worldbuilding ideation
- Making LLM calls from TouchDesigner
- Creative experimentations with generative narratives
- Demos and Showcase
Educational/Experiential Goals:
• Learn to write Python scripts, use APIs, and stream images and data in TouchDesigner
• Learn to build your own storytelling engine with TouchDesigner
• Make generative narratives
• Learn creative prompt engineering
• Explore creative ways to interact with a real-time image generation model
About Technologies:
• Fal.ai‘s real-time image generation APIs. We will be using the SDXL Turbo model, which can generate images quickly with just one step.
• OpenAI API, an API that provides access to the popular GPT language model.
• Socket.io, a library that enables low-latency, bidirectional and event-based communication between a client and a server.
• TouchDesigner
• Python
Accessibility Update:
Please note that this workshop will be held in the upstairs Gray Area classroom, which is not ADA accessible. We apologize for any inconvenience and encourage those impacted to contact us for assistance. Thank you for your understanding.
—
Become a Gray Area Member for early access to our workshops, and 20% off.
Gray Area Members enjoy extraordinary access to our interdisciplinary arts & education programs, while supporting our mission to catalyze creative action for social transformation. Get free admission, exclusive content, and more!
