Because they shipped and are better only for specific applications. For everything else, they lag behind frontier models. They win in latency-sensitive niches (e.g., autocomplete, code infilling, subagent calls). Not frontier-level reasoning.

They’re kinda dyslexic - like bad OCR