FlagshipMoonshot

Kimi K2.5

Kimi K2.5 is Moonshot's flagship text-only model with a 128K context window, designed for chat completion and code generation tasks.

Context 128K
Tier Flagship
Tools Supported
Input from
$0.383 / 1M tokens
across 3 providers

API Pricing

Cheapest on OpenRouter 23% below avg
ProviderInput / 1MOutput / 1MUpdated
$0.383$1.724/14/2026
$0.500$2.804/14/2026
$0.600$3.004/14/2026

Prices updated daily. Last check: 4/14/2026

Model Details

General

Creator
Moonshot
Family
Kimi
Tier
Flagship
Context Window
128K
Modalities
Text

Capabilities

Tool Calling
Yes
Open Source
No
Subtypes
Chat Completion, Code Generation
Aliases
kimi-k2-5-turbo, moonshotai-kimi-k2-5

Strengths & Limitations

  • 128K token context window supports long document processing
  • Tool calling functionality enables integration with external services
  • Flagship tier positioning within Moonshot's model lineup
  • Specialized focus on chat completion and code generation tasks
  • Available through multiple API aliases for flexible integration
  • Text-only design may offer performance advantages for language-focused tasks
  • No image or vision input capabilities
  • Proprietary model with no open-source availability
  • Smaller context window compared to some competing flagship models
  • Limited to text modality while many peers offer multimodal support
  • Fewer modality options than comprehensive flagship alternatives

Key Features

128,000 token context window
Chat completion capabilities
Code generation and programming assistance
Tool calling with external service integration
Text-only processing and generation
Multiple API endpoint aliases
Flagship-tier model architecture
Conversational AI optimization

About Kimi K2.5

Kimi K2.5 is the flagship model in Moonshot's Kimi family, representing the company's most capable offering for text-based AI tasks. As a proprietary model from the Chinese AI company Moonshot, it focuses on chat completion and code generation capabilities without multimodal features. The model operates with a 128,000 token context window and supports tool calling functionality. Kimi K2.5 is designed specifically for text-based interactions, emphasizing conversational AI and programming assistance. Unlike some contemporary flagship models that incorporate vision or other modalities, this model maintains a focused approach on text processing and generation. Kimi K2.5 serves users who need a capable text model for complex conversations, coding assistance, and scenarios requiring substantial context retention. While positioned as Moonshot's flagship offering, it competes in a market where other flagship models often provide larger context windows or multimodal capabilities, making it particularly relevant for users prioritizing Moonshot's specific approach to language modeling and tool integration.

Common Use Cases

Kimi K2.5 is well-suited for applications requiring sophisticated text processing and generation, particularly in conversational AI scenarios and software development contexts. Its 128K context window makes it effective for analyzing long documents, maintaining extended conversations, or working with substantial codebases. The tool calling capability enables integration into agentic workflows where the model needs to interact with external APIs or services. Given its flagship positioning and code generation focus, it works well for complex programming tasks, technical documentation, content creation, and scenarios where users need a capable text model without multimodal requirements. The model's specialization in chat completion makes it particularly relevant for customer service applications, virtual assistants, and interactive AI systems that prioritize conversational quality over multimodal features.

Frequently Asked Questions

How much does Kimi K2.5 cost per million tokens?

Kimi K2.5 pricing varies by provider and may include different rates for input and output tokens. Check the pricing table above for current rates across all available providers.

What is Kimi K2.5 best used for?

Kimi K2.5 excels at chat completion and code generation tasks, making it ideal for conversational AI applications, programming assistance, and scenarios requiring long-context text processing with its 128K token window. The tool calling feature also makes it suitable for agentic workflows requiring external service integration.

Does Kimi K2.5 support image or multimodal inputs?

No, Kimi K2.5 is a text-only model that focuses specifically on chat completion and code generation. It does not support image processing, vision capabilities, or other modalities beyond text input and output.