MiniMax Agent Team: built for long-running and continuous evolution
Original title:MiniMax Agent Team: Built for Long-Running Tasks and Continuous Evolution
AI summary
MiniMax upgraded its Agent Team overall and renamed it Mavis. The core of this update is the launch of Agent Teams, allowing users to run multiple agents with different roles in parallel on the MiniMax Agent desktop, forming teams to handle complex long-duration tasks that a single agent struggles to complete. Meanwhile, the original TokenPlan and Agent Plan merged into a unified subscription integrating CLI, API and agent access to the M2.7 model, music, video and voice features, with credits shared between agents and API. The upgrade aims to solve the stagnation, ambiguous judgment and quality degradation that single agents encounter in long tasks, delivering a smoother experience through a mechanism in which the lead agent responds quickly, tasks are split and executed in parallel, and key milestones are reported.
Key Takeaways
- MiniMax upgraded its Agent Team overall and renamed it Mavis.
- The core of this update is the launch of Agent Teams,allowing users to run multiple agents with…
- Meanwhile,the original TokenPlan and Agent Plan merged into a unified subscription integrating…
- 7 model,music,video and voice features,with credits shared between agents and API.
- The upgrade aims to solve the stagnation,ambiguous judgment and quality degradation that single…
Quick Facts
- Published
- Updated
- Source
- MiniMax:Blog(网页)
- Category
- Products & Apps
- AI score
- 75
MiniMax 把 Agent 从单兵作战升级成团队协作,解决了长任务里「Agent 变蠢」和「为什么没理我」两大痛点,做复杂任务的可以认真看看。
Related entities
Source & copyright
This article is based on curated titles and AI summaries from the AIHOT public API, with attribution and links back. Original copyright remains with the original publishers and authors; this site does not mirror full texts.
How to Cite
When citing this article, keep the title, date, and permalink, and credit Orange News as the source.
Orange News: MiniMax Agent Team: built for long-running and continuous evolution. May 27, 2026. https://new.nat6.net/en/news/cmpunu0g50003slbr6n7p2gqa/