Same world-gen prompt: Gemini 4 Argon clearly beats current 3.8 Flash
testingcatalog · x · 2026-10-12
A follow-up from testingcatalog: running the same world-gen prompt against the model served as "Gemini 3.8 Flash" vs the suspected Gemini 4 Argon shows a clear quality gap, offering a side-by-side look at the unreleased model. Consistent with the author's main leak post and blog writeup.
Related event: Gemini 4 Argon Reportedly A/B Testing in Antigravity(4 posts)→
More from Models
- GPT-6.1 Sol Matches Flagship Within 1 Point at 1/5 the Price; GPT-6.1 Astra Pulled Over Deception — DeepLearningAI · 2026-10-13
- AI memory tested: GPT-4o drops from 87% to 60.6% when fed the full 115k-token chat history — gaganghotra_ · 2026-10-13
- MoonMapBench: Opus 5.5 Off by 1,192 m Predicting Lunar Elevations — maksym_andr · 2026-10-13
- GPT-6.1 Sol Praised as Fast Workhorse, But Daybreak Mode Blocks Manual Model Picks — rez0__ · 2026-10-13
- ArchitectureQ benchmark: frontier LLMs hit 76% on training intuition vs 66% top humans — gregmushen · 2026-10-13
- Dev shares Claude Code bill: 469M tokens, $507 cost, 95.8% cache hit rate — PratikKadam_ · 2026-10-13