🎉 New kimi k2.5 Multi-modal Model released! Now supports multimodal understanding and processing.
Docs
Welcome

Welcome to Kimi API Docs

Kimi K2.5 Model

Kimi K2.5 is Kimi's most versatile model to date, featuring a native multimodal architecture that supports both visual and text input, thinking and non-thinking modes, and dialogue and agent tasks.

Quick Start

Getting Started with Kimi K2.5
Kimi K2.5 quick start guide and basic concepts
Console
Manage API keys, usage and billing
Getting Started with Kimi K2
Quickly understand and get started with Kimi K2 model
Pricing & Billing
Learn about model pricing and billing methods
Using Thinking Models
Learn about how to use Kimi model's thinking capabilities
FAQ
Find answers to common questions

Models and Pricing

View all
kimi-k2.5
Multimodal understanding
kimi-k2-turbo-preview
preview
kimi-k2-thinking
long think

More Resources

K2 Model Details
Learn about K2 model features and advantages
Development Workbench
Interactive interface to experience Kimi model capabilities
Web Search Tool
Use Kimi's official web search tool
Best Practices
Prompt writing and application examples
User Manual