General Reasoning Unveils Avalon-M1-Mini: 120B AI Model Trains Team-Based Reasoning With RLCR Method
According to Beating, General Reasoning has unveiled Avalon-M1-Mini, a 120 billion parameter model introducing RLCR (Collective Reward Reinforcement Learning), an approach that trains multi-agent teams rather than individual agents. The model deploys up to 8 agent employees collaborating via a
GateNews·10-25 06:05
