Replit launches Free Mode coding mode powered by OpenAI GPT-5.6 Luna
On August 19, Replit launched Free Mode inside Core and Pro subscriptions, powered by OpenAI's cheapest GPT-5.6 Luna. Core users get 30x more creation and 30 hours of monthly chat.
On August 19, AI coding platform Replit announced Free Mode inside its Core and Pro subscriptions, powered by OpenAI's GPT-5.6 Luna model for everyday chats, ideation, and lightweight Agent tasks without consuming the user's normal AI credits. Core stays at $20/month and Pro at $100/month.
Free Mode became feasible thanks to GPT-5.6 Luna's low pricing. OpenAI cut the model's price by roughly 80% on July 30, with input and output around $0.10 and $1.20 per million tokens (varies by region). With Luna, Replit can now absorb heavy usage while keeping subscription margins positive. Core users can now create 30 times more and receive 30 hours of monthly chat.
Free Mode fits everyday coding, simple edits, debugging, and proof-of-concept work. When complexity rises, Replit's Agent can auto-route requests to Power Mode or Max Mode, which use higher-cost or higher-performance models for large code changes, deep debugging, and long-running builds. OpenAI CEO Sam Altman said Free Mode is a concrete example of what better model economics can unlock. Replit founder and CEO Amjad Masad expects the partnership to expand the population able to build software by roughly 100x.
Replit and OpenAI have worked together since GPT-3 era. Free Mode is the first of "many launches" they plan to do together.