Kimi Agent Mode Kicks Off Beta Testing: AI Tackles Hours of Complex Tasks Autonomously
5 day ago / Read about 0 minute
Author:小编   

On September 26, Moonshot AI, a prominent domestic large - model company, launched a small - scale internal beta test for the Kimi Agent mode, dubbed "OK Computer." Some early adopters have already been granted access to try it out via the web - based Kimi platform. This step represents a significant upgrade for Kimi, transforming it from a mere conversational AI assistant into an intelligent agent capable of independently accomplishing complex tasks. Its core competitive edge lies in its continuous execution capability, autonomous planning skills, and the ability to leverage a wide array of tools.

With this mode, Kimi can operate for extended durations in a virtual computer environment to fulfill intricate instructions. This is made possible by a native Agent model that has been trained using end - to - end reinforcement learning technology. The AI can proficiently handle various task scenarios by making use of over 20 tools, such as file systems and browsers.