Developers, we’re previewing Reinforcement Fine-Tuning (RFT) for @OpenAI o1! With just a handful of examples, you can guide o1 to tackle complex, domain-specific tasks, and it works incredibly well. Rolling out early 2025—apply now for early access: https://
openai.com/form/rft-resea
rch-program/
…

→ View original post on X — @romainhuet,