Back

Unpopular opinion but idk why GPT-5 becomes dumb than GPT-4 (last one) It is giving more AI-sh answers.. it's only me?

I made a web version of my childhood game: Slate Journey. My friend and I used to play this game on a slate (small blackboard). Try it; it's challenging and fun! Play here: https://maku.fun/slate-journey

𝗔𝗽𝗽𝗹𝗲 𝗺𝗶𝗴𝗵𝘁 𝗷𝘂𝘀𝘁 𝗯𝗲 𝗽𝗹𝗮𝘆𝗶𝗻𝗴 𝟰𝗗 𝗰𝗵𝗲𝘀𝘀 𝗶𝗻 𝘁𝗵𝗲 𝗔𝗜 𝗴𝗮𝗺𝗲—𝗮𝗻𝗱 𝗵𝗲𝗿𝗲’𝘀 𝘄𝗵𝘆: 𝗢𝗽𝗲𝗻𝗔𝗜: GPT-4 → o3 → GPT-4 mini 𝗚𝗼𝗼𝗴𝗹𝗲: Gemini 1.5 → Gemini 2.5 𝗔𝗽𝗽𝗹𝗲: iOS 18 → iOS 26 — an 8-version leap ove

So Deepseek is nothing but an improved GPT4 wrapper. I'm not saying that's its bad, bust it's most likely a government funded company which burning as crazy (as GPT4 api costs are 107 times more than of deepseek v3)

Tried it, seems like downgraded version of GPT 3.5, still pretty impressive. I am sure but I guess they are not using their own llm like GPT, are they using GPT's API and base for krutrim Now, imagine yourself as bhavish Agrawal. If you had all co

There are 4 web apps on the development pipeline (One is frenzi: similar to instagram) nothing new in that except for the content moderation service, algorithm & monetization.

Hello Guys 🤩, Claude Opus model is dominating the AI race with 86.8% by competing with GPT-4 with 256 K token processing power and GPT-4 with 138 K tokens . Rumor Is that GPT-5 will launch in August with 1 Million tokens , isn't it interesting 😲

GPT-4 has outperformed human financial analysts in predicting earnings growth, and the finance world is abuzz. - With 60% accuracy, GPT-4 surpasses human analysts who manage a modest 53%-57%. Robots: 1, Humans: 0. - GPT-4's vast knowledge base and