Hands-on Gemini 3.8 Flash: Rebuilds a Mobile Game in 35 Seconds, Matches Opus 5 at Fraction of Cost
WorldofAI · youtube · 2026-09-05
- YouTuber WorldofAI ran a full test of Google's newly released Gemini 3.8 Flash across coding, reasoning, agentic workflows, Three.js simulations, game dev, and long-horizon tasks.
- Highlights: extremely fast and cheap; the model rebuilds a mobile mini-game in 35 seconds and shows big one-shot game-gen improvements over Gemini 3.7 Flash, while approaching Claude Opus 5 performance at a fraction of the cost.
- The interesting wrinkle: real-world coding quality vs GPT-5.6 Sol is much closer than benchmark claims suggest, and the video digs into where each wins.
- Official Google blog and multiple third-party test links are included for cross-checking.
More from Models
- 105 Real Bugs Tested: GPT-6.1 Sol Over 10x Cheaper Than GPT-5.6 Sol — PawelHuryn · 2026-10-23
- Abliterated V4.1-Flash eval: abstract algebra drops 6pp but college math gains 5pp — teortaxesTex · 2026-10-23
- Claude users hit "model at capacity" errors hourly, canceling plans over it — Alex-Stack · 2026-10-23
- Kimi K3.1 launch imminent: 3T-param bet on RL seen as make-or-break — teortaxesTex · 2026-10-23
- Uncensored fine-tunes vanish from Hugging Face amid unexplained takedowns — umarmnaq · 2026-10-23
- Claude app menu leaks mystery model name 'Fable' as user hits limits — ezshine · 2026-10-23