Kimi K2.6 Pricing, Benchmarks, Latency & Providers

Providers

API providers, route pricing, availability, and recent reliability signals.

Performance

Latency, throughput, and reliability signals from recent traffic.

Pricing

Effective prices over the last 30 days, with current provider list prices for context.

Benchmarks

Headline benchmark standings and comparison context.

Activity

Daily gateway activity over the last 30 days, with current UTC-day pace projection.

Apps Using This Model

Public apps observed in gateway request traffic for this model.

Model Uptime

Uptime trend for this model over the last 24 hours.

Quickstart

Start calling this model with endpoint-specific examples.

About

Key dates, capabilities, and model metadata.

Subscriptions

Commercial plans and bundled access that currently include this model.

import AIStats from '@ai-stats/sdk'; const client = new AIStats({ apiKey: process.env.AI_STATS_API_KEY, }); const response = await client.generateResponse({ "model": "moonshotai/kimi-k2.6", "input": "Give me one fun fact about cURL." }); const outputText = response.output ?.flatMap((item) => item.content ?? []) .find((item) => item.type === "output_text") ?.text; console.log(outputText ?? response);

Parameter	Description
`temperature`	Controls how random token selection can be.
`top_p`	Applies nucleus sampling by limiting candidates to a probability mass threshold.
`top_k`	Restricts sampling to the top-k candidate tokens on providers that expose it.
`max_tokens`	Caps output length on endpoints and providers that use the max_tokens field name.
`frequency_penalty`	Discourages repeated tokens in proportion to how often they already appeared.
`presence_penalty`	Encourages the model to explore new wording or topics after they first appear.
`repetition_penalty`	Applies provider-specific anti-repetition behavior outside the classic penalty fields.
`seed`	Requests deterministic sampling when the upstream provider supports seeded generation.
`stop`	Defines one or more sequences that terminate generation early.
`logprobs`	Requests token-level probability data in the response.
`tools`	Defines callable tools or functions the model can invoke.
`structured_outputs`	Capability signal for reliable schema-constrained output workflows.
`include_reasoning`	Requests reasoning content or reasoning summaries in responses where supported.
`logit_bias`	Adjusts token selection bias directly when a provider exposes logit control.
`min_p`	Narrows sampling by discarding tokens below a minimum probability threshold.
`top_logprobs`	Limits how many alternative token probabilities are returned per position.

Parameter

Description

temperature

Controls how random token selection can be.

top_p

Applies nucleus sampling by limiting candidates to a probability mass threshold.

top_k

Restricts sampling to the top-k candidate tokens on providers that expose it.

max_tokens

Caps output length on endpoints and providers that use the max_tokens field name.

frequency_penalty

Discourages repeated tokens in proportion to how often they already appeared.