<!-- mobian-agent-page publisher="dailydev" canonical="https://daily.dev/posts/openjiuwen-s-agent-swarm-when-ai-agents-finally-work-as-a-team-pffvehhee" -->

---
title: openJiuwen’s Agent Swarm: When AI Agents Finally Work as...
description: openJiuwen is an open-source AI agent platform built by Huawei Laboratories, Huawei Cloud, and the Terminal Xiaoyi team, released under Apache 2.0. Its...
canonical: https://daily.dev/posts/openjiuwen-s-agent-swarm-when-ai-agents-finally-work-as-a-team-pffvehhee
twitter:card: summary_large_image
twitter:site: @dailydotdev
og:type: website
og:site_name: daily.dev
og:title: openJiuwen’s Agent Swarm: When AI Agents Finally Work as a Team | daily.dev
og:description: openJiuwen is an open-source AI agent platform built by Huawei Laboratories, Huawei Cloud, and the Terminal Xiaoyi team, released under Apache 2.0. Its...
og:url: https://daily.dev/posts/openjiuwen-s-agent-swarm-when-ai-agents-finally-work-as-a-team-pffvehhee
og:image: https://api.daily.dev/og/posts/PFfVeHheE.png
og:image:alt: openJiuwen’s Agent Swarm: When AI Agents Finally Work as a Team
og:image:width: 1200
og:image:height: 630
og:locale: en
---

> ## Documentation Index
> Fetch the complete documentation index at: https://daily.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# openJiuwen’s Agent Swarm: When AI Agents Finally Work as a Team

**[It's Foss](https://daily.dev/sources/itsfoss)** · 4 min read · 1 upvotes · 0 comments

## Summary

openJiuwen is an open-source AI agent platform built by Huawei Laboratories, Huawei Cloud, and the Terminal Xiaoyi team, released under Apache 2.0. Its flagship component, JiuwenSwarm, implements "Coordination Engineering" — enabling multiple AI agents to autonomously divide labor, dynamically negotiate, and collaborate on complex tasks. Key features include heterogeneous model routing (assigning heavy-reasoning models to complex tasks and lightweight models to simpler ones), a layered persistent memory architecture with cross-swarm experience inheritance, distributed deployment across different nodes and servers, and an observability layer with self-evolving task templates. JiuwenSwarm integrates natively with HarmonyOS PC, allowing phones and tablets to act as distributed agent nodes. It is already in use by OfficeClaw for enterprise office automation and is open for community contributions on GitHub.

## Full article

daily.dev links to this article rather than hosting it. Read it at the original source: <https://feed.itsfoss.com/link/24361/17411448/openjiuwens-agent-swarm>

## Questions this post answers

### What is JiuwenSwarm and how does it differ from a single AI agent approach?

JiuwenSwarm is an open-source multi-agent orchestration framework released by the openJiuwen project (built by Huawei Laboratories, Huawei Cloud, and the Terminal Xiaoyi team) in May 2026 under Apache 2.0. Instead of building one all-powerful agent, it lets multiple agents autonomously divide labor, negotiate, and collaborate, using heterogeneous model routing, layered memory, and distributed deployment across processes, nodes, or servers.

_Developers weighing single-agent versus multi-agent architectures can track frameworks like JiuwenSwarm on daily.dev._

### How does JiuwenSwarm integrate with HarmonyOS PC for multi-device agent collaboration?

In the HarmonyOS PC Agent PC environment, the PC serves as the home base for the Leader Agent while HarmonyOS phones and tablets act as Teammate Agent nodes, working together without extra adaptation. This distributed architecture also supports the Leader Agent and swarm members running across different processes, nodes, or servers, useful for tasks needing isolated environments like intranet databases.

_Anyone building cross-device agent apps on HarmonyOS can follow this kind of integration work on daily.dev._

### How does JiuwenSwarm assign different AI models to different agent roles in a swarm?

JiuwenSwarm supports heterogeneous model routing, letting developers declare a role-to-model mapping in a configuration file so the system automatically allocates heavy-reasoning models to complex logic tasks and lightweight models to simple tasks. This reduces system load and improves overall execution quality, and it already integrates with platforms like HarmonyOS Xiaoyi and Lark (Feishu).

_Teams designing cost-efficient multi-model agent pipelines can follow developments like this on daily.dev._

## Similar posts on daily.dev

- [OpenClaw fever grips China](https://daily.dev/posts/openclaw-fever-grips-china-pardujuxo) · TechCentral · 11 upvotes · 4 comments
- [Researchers are tracking a Chinese AI 'agent fleet'](https://daily.dev/posts/researchers-are-tracking-a-chinese-ai-agent-fleet--a7gcrryeq) · TechCrunch · 2 upvotes · 0 comments
- [Hustlers are cashing in on China’s OpenClaw AI craze](https://daily.dev/posts/hustlers-are-cashing-in-on-china-s-openclaw-ai-craze-i1v7czr2y) · MIT Technology Review · 1 upvotes · 0 comments
- [Tongyi DeepResearch: A New Era of Open-Source AI Researchers](https://daily.dev/posts/tongyi-deepresearch-a-new-era-of-open-source-ai-researchers-1nj7zb0o6) · Hacker News · 2 upvotes · 0 comments

---

Tags: [#open-source](https://daily.dev/tags/open-source), [#ai-agents](https://daily.dev/tags/ai-agents)

[View this post on daily.dev](https://daily.dev/posts/openjiuwen-s-agent-swarm-when-ai-agents-finally-work-as-a-team-pffvehhee)

```json
{"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]}
{"@context":"https://schema.org","@type":"TechArticle","headline":"openJiuwen’s Agent Swarm: When AI Agents Finally Work as a Team","url":"https://daily.dev/posts/openjiuwen-s-agent-swarm-when-ai-agents-finally-work-as-a-team-pffvehhee","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/openjiuwen-s-agent-swarm-when-ai-agents-finally-work-as-a-team-pffvehhee"},"datePublished":"2026-08-10T13:02:17.477Z","dateModified":"2026-09-13T18:30:55.748Z","description":"openJiuwen is an open-source AI agent platform built by Huawei Laboratories, Huawei Cloud, and the Terminal Xiaoyi team, released under Apache 2.0. Its...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/204f2d68b9ba25e7119de27e77ce9f8c?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/204f2d68b9ba25e7119de27e77ce9f8c?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"It's Foss","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"It's Foss","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/582e3faed71842bfb9d3840a787d6700","url":"https://daily.dev/sources/itsfoss"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/openjiuwen-s-agent-swarm-when-ai-agents-finally-work-as-a-team-pffvehhee","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"open-source,ai-agents","timeRequired":"PT4M"}
{"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"It's Foss","item":"https://daily.dev/sources/itsfoss"},{"@type":"ListItem","position":3,"name":"openJiuwen’s Agent Swarm: When AI Agents Finally Work as a Team"}]}
{"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/openjiuwen-s-agent-swarm-when-ai-agents-finally-work-as-a-team-pffvehhee#faq","mainEntity":[{"@type":"Question","name":"What is JiuwenSwarm and how does it differ from a single AI agent approach?","acceptedAnswer":{"@type":"Answer","text":"JiuwenSwarm is an open-source multi-agent orchestration framework released by the openJiuwen project (built by Huawei Laboratories, Huawei Cloud, and the Terminal Xiaoyi team) in May 2026 under Apache 2.0. Instead of building one all-powerful agent, it lets multiple agents autonomously divide labor, negotiate, and collaborate, using heterogeneous model routing, layered memory, and distributed deployment across processes, nodes, or servers. Developers weighing single-agent versus multi-agent architectures can track frameworks like JiuwenSwarm on daily.dev."}},{"@type":"Question","name":"How does JiuwenSwarm integrate with HarmonyOS PC for multi-device agent collaboration?","acceptedAnswer":{"@type":"Answer","text":"In the HarmonyOS PC Agent PC environment, the PC serves as the home base for the Leader Agent while HarmonyOS phones and tablets act as Teammate Agent nodes, working together without extra adaptation. This distributed architecture also supports the Leader Agent and swarm members running across different processes, nodes, or servers, useful for tasks needing isolated environments like intranet databases. Anyone building cross-device agent apps on HarmonyOS can follow this kind of integration work on daily.dev."}},{"@type":"Question","name":"How does JiuwenSwarm assign different AI models to different agent roles in a swarm?","acceptedAnswer":{"@type":"Answer","text":"JiuwenSwarm supports heterogeneous model routing, letting developers declare a role-to-model mapping in a configuration file so the system automatically allocates heavy-reasoning models to complex logic tasks and lightweight models to simple tasks. This reduces system load and improves overall execution quality, and it already integrates with platforms like HarmonyOS Xiaoyi and Lark (Feishu). Teams designing cost-efficient multi-model agent pipelines can follow developments like this on daily.dev."}}]}
```

