moonshotai/kimi-k2.6

OfficialView on Replicate →

Moonshot AI's frontier open model, built for long-horizon coding, agent swarms, and autonomous software engineering. 1 trillion parameters, 262k context window, vision and tool use.

Capabilities

Reference ImagesSystem PromptMax TokensTop-P

Cost

Community model (estimated from hardware time)

Input Parameters

Name	Type	Description	Default	Constraints
`frequency_penalty`	number	Penalizes new tokens based on their existing frequency in the text so far.	`0`	min: -2, max: 2
`image_input`	array	Optional list of images to include in the prompt. Kimi K2.6 supports vision inputs.		—
`max_tokens`	integer	Maximum number of tokens to generate.	`2048`	min: 1, max: 32768
`presence_penalty`	number	Penalizes new tokens based on whether they appear in the text so far.	`0`	min: -2, max: 2
`prompt`	string	Text prompt to send to Kimi K2.6.	`""`	—
`reasoning_effort`	string	How much effort the model spends reasoning before answering. Use 'none' to disable.	`"none"`	nonelowmediumhigh
`system_prompt`	string	System prompt to guide the model's behavior.	`"You are a helpful assistant."`	—
`temperature`	number	Sampling temperature. Higher values make output more random.	`0.6`	min: 0, max: 2
`top_p`	number	Nucleus sampling probability threshold.	`1`	min: 0, max: 1

frequency_penaltynumber

Penalizes new tokens based on their existing frequency in the text so far.

Default: 0min: -2, max: 2

image_inputarray

Optional list of images to include in the prompt. Kimi K2.6 supports vision inputs.

Default:

max_tokensinteger

Maximum number of tokens to generate.

Default: 2048min: 1, max: 32768

presence_penaltynumber

Penalizes new tokens based on whether they appear in the text so far.

Default: 0min: -2, max: 2

promptstring

Text prompt to send to Kimi K2.6.

Default: ""

reasoning_effortstring

How much effort the model spends reasoning before answering. Use 'none' to disable.

Default: "none"

nonelowmediumhigh

system_promptstring

System prompt to guide the model's behavior.

Default: "You are a helpful assistant."

temperaturenumber

Sampling temperature. Higher values make output more random.

Default: 0.6min: 0, max: 2

top_pnumber

Nucleus sampling probability threshold.

Default: 1min: 0, max: 1

Version: a9d565c31037Updated: 8/1/202664.4K runs