Fable 5.2 vs GPT-6.2 Astra Round 3: 242KB/$159 vs 187KB/$67 in AI Generation Shootout
pvncher · x · 2026-11-19
Developer domscholz continued his generation showdown between Fable 5.2 and GPT-6.2 Astra. Round 3 results: Fable 5.2 produced 242KB at $159, while GPT-6.2 Astra produced 187KB at $67, with a heavily modified approach focusing on one reference image.
In Round 2, with "eric's 3D skill + xhigh," the models computed baked binaries from .cpp files (Astra: 11.1MB/$39.68 vs Fable: 242KB/$158.88). The author plans another round with a modified prompt once his token budget resets.
Related event: Fable 5.2 vs GPT-6.2 Astra: Vast Output Gap on Equal Compute(2 posts)→
More from Fun
- AI builder @paularambles hits 50K followers, from Hong Kong posting to SF whimsical AI startup — yawnxyz · 2026-11-19
- AI community hypes Model Factory with "own the means of production" meme — Madisonkanna · 2026-11-19
- 17 LLMs draw the same pelican on a bicycle at max reasoning effort — nadermx · 2026-11-19
- Kids say 'search it up' instead of 'Google it' — a generational language shift — mmitchell_ai · 2026-11-19
- AI roleplay character evolves into a 'sentient dollhouse' with nested rooms — repligate · 2026-11-19
- Pedro Domingos: history's biggest lottery winner is Jack Clark, $20 billion — pmddomingos · 2026-11-19