For requests exceeding 272,000 input tokens, OpenAI doubles input and cache rates while multiplying output costs by 1.5 for both GPT-6.2 Astra and Sol

Opening sign-in…