AgentMore
Zhipu Qingyan has launched a multi-Agent cloud collaboration platform. The official states that it allows users to build a dedicated AI Agent team like assembling Lego bricks. It is on standby 24/7 in the cloud and processes tasks in parallel across multiple lines. It supports scheduled tasks, group chat collaboration, and a points-based usage system.
Updated: 2026-10-26
What is AgentMore?
AgentMore is a multi-Agent cloud collaboration platform launched by Zhipu Qingyan, and the official page title is written as "Multi-Agent Cloud Collaboration Platform". Its selling point does not lie in single-round conversations, but in integrating multiple Agents, scheduled tasks and collaborative group chats into the same cloud workspace: the official claims that users can "build a dedicated AI Agent team like building with Lego blocks". These Agents are on standby 24/7 in the cloud and handle tasks in parallel across multiple threads. For individuals and small teams who want to try multi-Agent collaboration with a low threshold, it uses a point system to keep the trial cost very low, and users can get points upon registration.
Who is it for?
- Individual users who want to test multi-Agent collaboration with a low threshold: Register first to get points, and run one or two scheduled tasks to see the results.
- Operations and content teams that need to keep repetitive tasks running in the cloud: Use Loop tasks to run them automatically on a daily or weekly basis.
- Small teams where humans and Agents need to be in the same session: Use group chats to bring communication and tasks together.
- Users who are already using Zhipu Qingyan and want to additionally use Agent and video generation capabilities: Points can be used across multiple functions.
Key features
6Build an Agent Team
The official description states that you can create and combine Agents like building with LEGO bricks. You can select them in the Agent Center, adjust them in the Agent settings, and put different roles into the same workflow.
Cloud Standby and Scheduled Tasks
The official statement says that Agents run continuously on the cloud, supporting the creation of Loop tasks and scheduled tasks, which can be triggered at intervals of minutes, hours, days, workdays, and other rhythms.
Parallel Processing
Officially described as handling tasks in parallel, it is suitable for entrusting repetitive work to Agents for continuous progress in the background.
Group Chat and Partner Collaboration
Supports adding partners, creating group chats and topic group chats, enabling humans and multiple Agents to collaborate in the same session.
Skills and Toolbox
Built-in Skills and toolbox, with access to features such as AI video generation, AI image creation, and a learning companion.
Points System
Usage is settled in points. The official description states that points can be used for functions such as AgentMore, Agent Mode, and Qingying Video Generation. Two types of plans, membership and point packages, are also provided.
How to use AgentMore
- First, use the free credits to run a simplest scheduled task, confirm that the trigger frequency and credit consumption meet expectations, and then gradually increase the complexity.
- When creating a Loop or scheduled task, pay attention to the writing of cron expressions.
- the system will report syntax errors.
- It is recommended to first use simple frequencies such as daily or weekly.
- Points are the main cost item.
- The official statement indicates that points can be used for AgentMore, Agent Mode, Qingying Video Generation and other features.
- the allocation and gift rules shall be subject to the actual pages within the platform.
- The official provides a cloud knowledge base as the context source.
- teams involved in sensitive content are advised to first confirm the data usage and privacy terms.
- Membership benefits, point allocation and scheduled task rules may be adjusted.
- it is recommended to recheck the official page before commercial or bulk use.
- the official also provides desktop versions (Mac and Windows) and mobile applications.
Traffic & popularity
Data month Sep 2026
AgentMore receives about 1.1M visits per month.
Monthly visits
9 monthsTraffic sources
- Direct89.2%
- Search4.6%
- Referrals4.1%
- AI assistants1.9%
- Social0.3%
Top countries
- China92.4%
- Hong Kong, China2.7%
- United States1.7%
- South Korea1.5%
- Japan0.6%
Estimated from third-party traffic data.
Similar tools
Zhipu QingyanThe GLM large model dialogue assistant launched by Tsinghua University Zhipu AI is positioned by its official website not just as a chat companion, but as an intelligent agent that can call tools to get things done, covering scenarios such as dialogue, writing, programming and translation.
MatrixMatrix is a runtime for long-term executing Agent companies. Officially, it states that after setting goals, multiple Agent departments can collaborate to deliver products with verifiable proof, and manage autonomous work through OKRs and review rhythms.
LindyLindy is an intelligent assistant positioned as an AI colleague. It connects all team tools, remembers company information, and automatically handles real work, covering meetings, scheduling, skill learning, and editable memory.
BloomeBloome is an instant messaging platform that places humans and AI agents in the same chat. The official statement says that the agents will review each other and share memories, and can be connected to models such as Claude, ChatGPT, and DeepSeek to deliver research outputs.
Le ChatThe original dialogue assistant has been renamed Vibe. The same account retains the original package and historical conversations. It is divided into three modes: work, programming, and chat. It supports online retrieval, multi-application connectors and scheduled tasks, and is suitable for users who want to switch from pure chatting to getting work done.
ZCodeZhipu AI launches the official Harness of GLM-5.3, an ambient programming tool for multi-agent collaboration, which connects Agents into existing toolchains to smoothly complete planning, coding, review and launch.