I use LLMs for coding and I would instantly notice. It’s GPT4 or Claude, Gemini a close third, llama and rest are far away. The harder the question, the better

I use LLMs for coding and I would instantly notice. It’s GPT4 or Claude, Gemini a close third, llama and rest are far away. The harder the question, the better OpenAI performs