How do I choose between GPT-4, Claude, Gemini, and open-source models?

Consider: task type (coding, reasoning, vision), context window size, latency requirements, cost per token, data privacy (cloud vs self-hosted), and model licensing. Open-source models (Llama, Mistral, Qwen) are preferable when data must stay on-premise. Benchmark on your specific workload leaderboard scores rarely translate directly.