PulseAugur
EN
LIVE 08:15:43

Kimi K2.5 multimodal agent model released with Agent Swarm framework

Researchers have introduced Kimi K2.5, an open-source multimodal agentic model designed to enhance general agentic intelligence through joint optimization of text and vision. The model incorporates techniques like joint text-vision pre-training and reinforcement learning. Kimi K2.5 also features Agent Swarm, a framework for parallel agent orchestration that breaks down complex tasks into concurrent sub-problems, reducing latency by up to 4.5x. The model achieves state-of-the-art results in coding, vision, reasoning, and agentic tasks, with its checkpoint being released for further research and application. AI

IMPACT This release offers an open-source multimodal agent model and orchestration framework, potentially accelerating research and development in agentic AI.

RANK_REASON The item is a research paper detailing a new model and framework. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

Kimi K2.5 multimodal agent model released with Agent Swarm framework

COVERAGE [1]

  1. arXiv cs.AI TIER_1 Italiano(IT) · Kimi Team, Tongtong Bai, Yifan Bai, Yiping Bao, S. H. Cai, Yuan Cao, Ziwei Chai, Y. Charles, H. S. Che, Cheng Chen, Guanduo Chen, Huarong Chen, Jia Chen, Jianlong Chen, Jun Chen, Kefan Chen, Liang Chen, Ruijue Chen, Xinhao Chen, Yanru Chen, Yanxu Chen, Y… ·

    Kimi K2.5: Visual Agentic Intelligence

    arXiv:2602.02276v2 Announce Type: replace-cross Abstract: We introduce Kimi K2.5, an open-source multimodal agentic model designed to advance general agentic intelligence. K2.5 emphasizes the joint optimization of text and vision so that two modalities enhance each other. This in…