Kimi K2.5
VerifiedProcesses extended text and image inputs for in-depth multimodal analysis.
About Kimi K2.5
Kimi K2.5 was developed by Moonshot AI as a proprietary system without public weights. Its architecture supports simultaneous handling of textual and visual data streams. The design emphasizes stability across very long input sequences.
Strengths include reliable retention of information over hundreds of thousands of tokens and coherent interpretation of combined text-image prompts. It avoids reliance on external retrieval for context that fits inside its native window.
Typical usage covers detailed document review, multi-turn conversations with attached images, and structured analysis of mixed-media sources. Developers integrate it into applications needing sustained context without frequent summarization steps.
Capabilities
How Kimi K2.5 compares
Kimi K2.5 (striped bar) vs other multimodal on intelligence, speed and price.
Price
USD per 1M output tokens · Lower is better · Kimi K2.5 ranks #30 of 67
Nova 2 LiteSources: Artificial Analysis (intelligence, speed) · OpenRouter (price).
Best for
Extended Document Analysis
Kimi K2.5 processes documents spanning its full 262144-token context window, enabling coherent summarization and insight extraction from lengthy reports or research collections.
Multimodal Image-Text Tasks
The model performs vision understanding and multimodal text-image analysis, supporting queries that combine visual elements like charts or photos with surrounding textual explanations.
Long-Context Conversational Reasoning
It maintains reasoning quality across extended dialogues, referencing prior exchanges or multiple images without losing thread continuity.
Strengths & limitations
Strengths
- +Very large context window
- +Native image support
- +Strong at long-input tasks
Limitations
- –Text and image only
- –Less widely tested ecosystem
- –Newer model with evolving features
Cost calculator
Estimate what Kimi K2.5 would cost for your usage.
Based on Kimi K2.5's $0.38/1M input · $2.02/1M output. Estimate only — actual cost varies by provider and caching.
Quick start
OpenRouter's API is OpenAI-compatible — most SDKs work by just swapping the base URL. Only the model slug changes between models.
import OpenAI from "openai";
const client = new OpenAI({
baseURL: "https://openrouter.ai/api/v1",
apiKey: process.env.OPENROUTER_API_KEY,
});
const completion = await client.chat.completions.create({
model: "moonshotai/kimi-k2.5",
messages: [{ role: "user", content: "Hello!" }],
});
console.log(completion.choices[0].message.content);Model slug: moonshotai/kimi-k2.5
Editor's verdict
Kimi K2.5 is Moonshot AI's proprietary multimodal with a 262K-token context window.
At $2.02 per 1M output tokens, it is mid-priced for its class.
It is available through Moonshot AI's API and aggregators like OpenRouter.
Best suited to very large context window and native image support.
Frequently asked questions
Kimi K2.5 offers a context window of 262144 tokens.
User reviews
Real, verified reviews from the community shape this model's rating.
Loading reviews…
Other Kimi models
Sibling versions in the Kimi family from Moonshot AI.
Similar models
Other multimodal worth comparing.