⚠ This model is no longer available

Kimi K2.5 was removed from RealRelay on 2026-10-19. This page is kept as a historical record: prices mentioned below are no longer valid, and the API no longer serves this model.

Kimi K2.5

Moonshot AI's native multimodal model with text, image and video input, thinking and non-thinking modes, and agent support.

Similar models still available

Overview

Kimi K2.5 is a native multimodal model from Moonshot AI. It accepts text, image and video input and supports thinking and non-thinking modes, so it can be used for both dialogue and agent tasks. Moonshot AI points to its strengths in visual coding, agent work and visual understanding. It returns text.

Kimi K2.5 has a 256K-token context window and can generate up to 16K output tokens. It accepts text, image and video input and returns text output. It was released on January 27, 2026.

Capabilities

Kimi K2.5 supports reasoning, tool calling, MCP, prompt caching and web search. It accepts text, image and video as input and produces text as output.

Thinking mode handles multi-step reasoning, non-thinking mode answers quickly, and tool calling, MCP and web search let Kimi K2.5 act as an agent. Prompt caching lowers the price of repeated context.

Pricing

Kimi K2.5 costs $0.57 per 1M input tokens and $3.00 per 1M output tokens. Cache reads are $0.10 per 1M tokens.

Kimi K2.5 is available through RealRelay's OpenAI-compatible API.

Use cases

Use Kimi K2.5 for visual coding: give it a screenshot or short video of an interface and ask for code that reproduces it, with tool calling to run and check the result.

It also suits agent tasks that search and call tools, and question answering over long documents or images within its 256K-token context window.

What is Kimi K2.5?

Kimi K2.5 is a native multimodal model from Moonshot AI. It supports reasoning, tool calling, MCP, prompt caching and web search and accepts text, image and video input.

How much does Kimi K2.5 cost?

Kimi K2.5 costs $0.57 per 1M input tokens and $3.00 per 1M output tokens, with cache reads at $0.10 per 1M tokens.

What is the context window of Kimi K2.5?

Kimi K2.5 has a 256K-token context window and can generate up to 16K tokens in a single response.

What inputs and outputs does Kimi K2.5 support?

Kimi K2.5 accepts text, image and video input and returns text output.

Does Kimi K2.5 support tool calling?

Yes. Kimi K2.5 supports tool calling, so it can call functions you define and use the results in multi-step tasks. It also supports reasoning, MCP, prompt caching and web search.

When was Kimi K2.5 released?

Kimi K2.5 was released on January 27, 2026.

How do I call Kimi K2.5 via the API?

Call Kimi K2.5 through RealRelay's OpenAI-compatible API using the model ID jd/kimi-k2.5.

Pricing

no longer offered
Kimi K2.5 was removed from RealRelay on 2026-10-19. This page is kept as a historical record: prices mentioned below are no longer valid, and the API no longer serves this model.

Specifications (as of 2026-10-19)

Context window
256K
Max output
16K
Released
2026-01-27
Input
Image / Text / Video
Output
Text
Capabilities
MCP, Prompt caching, Reasoning, Tool use, Web search