I made @OpenAI GPT-5.5 and @GoogleDeepMind Gemini 3.1 Pro play REAL UNO — 10 games, full rules.
Gemini overthought every single card — literally 6× more reasoning tokens (133.5K vs 22.5K).
And it paid off — Gemini won 8–2.
Best part: they roast each other between cards the
This is Overchat: The All-AI-in-One chat!
No need to pay for separate tools — access ChatGPT, Claude, Gemini, Grok, and more in one super app.
Joined December 2024
- New Opus 4.8 crushed every frontier model at simulating water physics. One prompt to all four (GPT-5.5, Gemini 3.1 Pro, old Opus 4.7): a single HTML file, fluid that chases your cursor. Opus 4.8 delivered an unreal, hypnotic result — worthy of its own app — while burning the
- We asked Opus 4.8 to introduce itself. It made this. 👇 Claude's most powerful model is now live in Overchat → overchat.ai/models/claude/…
- Omni Flash is basically nano banana but for video the resolution isn't even the point, it's the physics veo renders a frame, but omni flash renders actual movement — weight, inertia, the foot hitting the roof you can tell they put in the work. well done google see for yourself
- New Gemini 3.5 Flash is cheaper and faster than the big models — and it just out-coded all of them. Same prompt to 4 AIs: "build Space Invaders game": - Flash shipped the only complete, polished game that actually runs. 1,020 lines. - Grok was faster but barebones. - Claude

