Designer tries making a typeface with VQGAN: "much harder than I thought"
makeitrad1 · x · 2026-09-22
The author attempted to create a typeface using VQGAN image generation and found it far harder than expected, managing only the letters A through D before sharing the rest of the alphabet in a thread. A hands-on test of generative models for type design.
Related event: Designer Tests VQGAN Font Creation, Finds It Harder Than Expected(2 posts)→
More from Multimodal
- Seeking a local/free workflow for 5-shot character-consistent video (Krea + H3 + ComfyUI) — NewPhoneWhotiz · 2026-09-22
- Qwen 2.1 CFG test: CFG 1.0 + 12 steps halves image generation time — FreeTheClanks · 2026-09-22
- Qwen-Image-2.1 Tops Open Text-to-Image Arena at 1228 pts, Just 11 Behind GPT-Image — arena · 2026-09-22
- MiniMax H3 Roundup: ControlNet Union 2.0, MoGe Camera-Path Node and New LoRAs — optimisticalish · 2026-09-22
- Maestro video character swap workflow: sliding windows auto-prompt frames — cocktailpeanut · 2026-09-22
- Qwen-Image-2.1 takes 3-6 min per image even on an RTX Pro 6000 — unfinishedmaze · 2026-09-22