Prime Intellect launches reinforcement learning fine-tuning service to support customized agents for AI native companies
Source: PANews
Time: 2025-09-10 15:34:05
PANews reported on September 10 that the decentralized AI protocol Prime Intellect releases reinforcement learning fine-tuning (RFT) service, providing full-stack RL infrastructure, helping AI native companies create and train customized agents. Services include rl-trainer, validator, and environment center, which supports training models in reinforcement learning environments. In addition, Prime Sandboxes is also launched to isolate Docker environments, designed for secure code execution and LLM benchmarking.
Previous news, the decentralized AI protocol Prime Intellect completed a $15 million financing, led by Founders Fund.