X @Sam Altman

RT Ahmed El-Kishky (@ahelkky)My first project at OpenAI involved teaching our models to reason and use tools by improving their competitive programming skills. Back then, GPT-4 struggled with even the simplest Codeforces problems, often oom-ing in the sandbox. It's incredible to see that just 2.5 years later, our models now achieve IOI gold-level performance! This amazing progress reflects the hard work of many, many people. Huge congrats to everyone involved! ...