OpenAI will completely shut down the fine-tuning API: Large models will fully shift to Prompt

CryptoWorld News reports that OpenAI has sent an official email announcing the complete shutdown of its developer-facing self-service fine-tuning service. Starting immediately, new users can no longer create fine-tuning tasks, and existing active users can only use the service until January 6, 2027. As for already deployed fine-tuned models, their inference services will be tied to the lifecycle of the underlying base models; when the base models are retired, the services will cease. The core reason given by OpenAI is that the next-generation base models (such as GPT-5.5) are already extremely powerful in following instructions and formatting. Compared to expensive fine-tuning, directly writing prompts combined with retrieval-augmented generation (RAG) is not only cheaper and faster but also sufficient to cover the vast majority of scenarios. Tech media Startup Fortune believes this is not merely a product cleanup but that OpenAI is shrinking its underlying model interface. In the past, startups often used low-cost self-service fine-tuning to create customized products for vertical industries. With this channel closed, developers can only use general prompt engineering, significantly increasing the difficulty for startup teams to build customized technical barriers around the OpenAI ecosystem.