Text Generation Model

Kimi K2.6

Kimi K2.6 is a text generation model from Moonshot AI with a 262,144-token context window, served via Deep Infra.

Start Building with Kimi K2.6 View All Models

Publisher

Moonshot

TypeText

Context Window262,144 tokens

ReleasedApril 2026

Input$0.75/MTok

Output$3.50/MTok

Provider

DeepInfra

LATEST

Try Kimi K2.6 →

About Kimi K2.6

Long-context text generation from Moonshot AI

Kimi K2.6 is a large language model developed by Moonshot AI, published under the identifier moonshotai/Kimi-K2.6 and made available through the Deep Infra provider. It is a chat-oriented text generation model released in April 2026, designed to handle long-form inputs and outputs within a context window of 262,144 tokens and a maximum response size of 16,384 tokens.

The model is well-suited for tasks that require processing or generating large volumes of text in a single session, such as document analysis, extended conversations, summarization of lengthy content, and multi-step reasoning over long inputs. Its 262K context window places it among models capable of ingesting book-length documents or large codebases in a single pass, making it a practical choice for developers building applications that depend on retaining extensive context.

Capabilities

What Kimi K2.6 supports

Long Context Window

Supports up to 262,144 tokens of context in a single session, enabling ingestion of book-length documents or large codebases without truncation.

Text Generation

Generates coherent, extended text responses in a chat format, with a maximum response size of 16,384 tokens per turn.

Extended Conversation

Maintains context across long multi-turn dialogues, retaining information from earlier in the conversation up to the 262K token limit.

Document Analysis

Processes and reasons over lengthy documents in a single pass, making it suitable for summarization, extraction, and question-answering on large texts.

Instruction Following

Responds to structured prompts and complex instructions as a chat-tuned model, supporting developer and end-user task automation workflows.

Ready to build with Kimi K2.6?

Get Started Free

FAQ

Common questions about Kimi K2.6

What is the context window size for Kimi K2.6?

Kimi K2.6 supports a context window of 262,144 tokens, allowing it to process very long documents or extended conversations in a single session.

What is the maximum response length Kimi K2.6 can produce?

The model has a maximum response size of 16,384 tokens per generation turn.

Who developed Kimi K2.6 and when was it released?

Kimi K2.6 was developed by Moonshot AI and released in April 2026. It is available on MindStudio via the Deep Infra provider.

Does Kimi K2.6 support image or video inputs?

Based on the available metadata, Kimi K2.6 does not have confirmed support for image or video analysis inputs. It is classified as a text generation model.

What is the pricing for using Kimi K2.6 on MindStudio?

Pricing information for Kimi K2.6 has not been published in the available metadata. You can check the MindStudio platform for current pricing details.

Resources