OpenAI
GPT-6.2 No Longer Caches the instructions Field
GPT-6.2 Astra and Sol, released 17 November, place the Responses API's top-level instructions field after the conversation, so it never reaches the prompt cache. In my runs the same eight-turn conversation with a 4,340-token system prompt cost $0.0176 on input with GPT-6.1 Sol and $0.0795 with GPT-6.2 Sol, and $0.0168 once the prompt moved to a developer message.
19 Nov 20266 min read