Reinforcement Fine-Tuning (RFT)

In depth

Fine-tuning a model with reinforcement learning against a grader you define, instead of on fixed correct answers. The training loop samples several responses per prompt, scores each one with the grader, and shifts the weights so high-scoring outputs become more likely. OpenAI offers RFT as a managed API for its reasoning models, and teams run the same pattern themselves with GRPO on open-weight models. It works best when quality can be scored reliably, because a weak grader invites reward hacking.

Primary source: platform.openai.com