See how different content sizes fill each model's context window. Know when you'll hit the limit before it matters.
Token counts are approximate. 1 token ≈ 4 chars of English prose, ≈ 3 chars of code. Context windows shown are standard API limits — some providers offer extended contexts.