# Prime Intellect Releases prime-rl 0.6.0 to Train Trillion-Parameter MoE Models on Agentic RL Workloads

来源：AI Canteen · AI 洞察
采集日期：2026-06-23
原始链接：https://www.marktechpost.com/2026/06/23/prime-intellect-releases-prime-rl-0-6-0-to-train-trillion-parameter-moe-models-on-agentic-rl-workloads/

---

Prime Intellect has releasedprime-rl version 0.6.0. The framework targets reinforcement learning on trillion-parameter Mixture-of-Experts (MoE) models. It focuses on heavy agentic workloads, like long-horizon software-engineering tasks.

---

© 2026 北京韩亚投资顾问有限公司 — AI Canteen