AI image generation has come far from making up gibberish text
fhinkel · x · 2026-09-26
Developer fhinkel notes how image generation models used to put slightly made-up words into images, and reflects on how far the technology has come since then.
More from Multimodal
- Uncensored 670M image model Nanosaur 2 fully released, quantizes down to 388MB — FusionCow · 2026-09-26
- Building a 3D game with GPT-6, Opus 5.5 and MCP for Blender — sidahuj · 2026-09-26
- Claude made a launch video by reading the repo — introducing Foleo, a local workspace for agent artifacts — joshgonsalves_ · 2026-09-26
- MiniMax H3 generates striking fashion photo shoot video, wowing Reddit — R34vspec · 2026-09-26
- For consistent faces in wide shots, render at 2.0MP — upscalers don't cut it — R34vspec · 2026-09-26
- $15k Launch Video Could've Been Made for $20, Teases AI User — zainhas · 2026-09-26