Comment by emaadm

OpenAI has had a fine-tuning API since GPT-3.5, and a reinforcement fine-tuning API since last year.