Back to News
Moonshot AIJanuary 27, 2026

Kimi K2.5

Open Source
LLMs
Agentic AI
AI Agents

Explore Kimi K2.5

Visit the official website to learn more and get started

### TL;DR

Kimi K2.5 is an advanced open-source visual agentic intelligence model developed by Moonshot AI. It integrates a large Mixture of Experts (MoE) language backbone with a native vision encoder and a parallel multi-agent system called Agent Swarm, enabling efficient execution of complex tasks through coordinated sub-agents.

Key Insights & Metrics

Pricing
Free
Cost structure
Version
Kimi K2.5
Current release version
Hardware
Compatible with standard GPUs; quantized INT4 variants available for deployment on commodity GPUs with lower memory budgets
Compute requirements
Category
Open Source
Licensing model
Region
China
Primary region

Key Features

  • Native multimodal training on 15 trillion mixed visual and text tokens
  • Agent Swarm system capable of managing up to 100 sub-agents executing 1,500 coordinated steps
  • Strong performance in coding, multimodal reasoning, and deep web research benchmarks

Related Releases

AIBuildAI

AIBuildAI is an AI agent that autonomously constructs AI models. Given a specific task, it initiates an agent loop to analyze the problem, design models, and execute training processes, all without human intervention.

AIBuildAI
Open

llmock

llmock is a deterministic mock LLM server designed for testing purposes. It provides a real HTTP server with authentic SSE streams, allowing developers to simulate interactions with various LLM APIs like OpenAI, Claude, and Gemini without incurring costs or dealing with non-deterministic results.

CopilotKit
Open

MiroThinker

MiroThinker is an open-source search agent model developed by MiroMindAI, designed for tool-augmented reasoning and real-world information seeking. It aims to match the deep research capabilities of leading AI models like OpenAI's Deep Research and Google's Gemini Deep Research.

MiroMindAI
Open

Letta Code SDK

The Letta Code SDK is a software development kit that enables developers to build deeply personalized agents with persistent memory that learn over time. It serves as the interface to Letta Code, facilitating the creation of stateful agents capable of continuous learning and improvement.

Letta AIJan 28
Open

Discussion

0
Upvotes
0
Downvotes
0 reviews

Sign in to leave a review

Reviews

No reviews yet. Be the first to review!

🚀 Join the AI dev community — follow us everywhere

© 2026 MARKTECHPOST AI MEDIA INC. All rights reserved.Terms & ConditionsPrivacy Policy
Beta Mode