Basics of Image Generation with ComfyUI
An introductory, practice-oriented session on how modern text-to-image systems work and how to build reproducible image-generation workflows in ComfyUI. The session covers core ideas behind Stable Diffusion, prompt design (including negative prompts), and hands-on workflow setup.
Format and structure
- Introduction: how AI image generation works; latent space, diffusion, prompts (15 minutes)
- ComfyUI essentials: install/launch, interface tour (10 minutes)
- Build your first workflow: model load, sampler, connections, generate (25 minutes)
- Improve results: advanced/negative prompts, resolution, img2img, control nodes (20 minutes)
- Tips & good practice: managing models/resources, debugging, saving/reusing (20 minutes)
- Guided hands-on + Q&A (20 minutes)
Who should attend: Students, designers, developers, founders, creators, and educators exploring AI-assisted visuals. No prior experience required.
Key objectives
- Understand the basics of diffusion-based image generation
- Navigate ComfyUI and construct a simple, reusable workflow
- Apply prompt techniques and sampler settings for higher-quality outputs
- Save, share, and iterate on workflows effectively
Preparation: Laptop with stable internet; ComfyUI pre-installed; basic familiarity with prompts.
Date: Thursday, September 18, 2025
Time: 4:00 - 6:00 PM IST
Venue: Online Via Zoom
