Kimi K2.6
Kimi K2.6 is Moonshot AI's natively multimodal flagship focused on long-horizon coding and design with code, with a context window of 262.1K tokens, available through AI Gateway via Moonshot AI, Fireworks, Novita AI, Baseten. Your use is subject to Moonshot AI's Terms & Privacy Policies.
View API reference- Input and output price
- Prices from: Input $0.95, Output $4, Per 1M tokens
- 24h uptime
- Loading AI Gateway uptime
import { streamText } from 'ai'
const result = streamText({ model: 'moonshotai/kimi-k2.6', prompt: 'Why is the sky blue?'})Copy link to headingPlayground
Try out Kimi K2.6 by Moonshot AI. Usage is billed to your team at API rates. Free users (those who haven't made a payment) get $5 of credits every 30 days.
Kimi K2.6
Kimi K2.6
Copy link to headingProviders
Route requests across multiple providers. Copy a provider slug to set your preference. Visit the docs for more info. Using a provider means you agree to their terms, listed under Legal.
Copy link to headingUptime24 hours
Direct request success rate on AI Gateway and per-provider. Visit the docs for more info.
Copy link to headingThroughput24 hours
P50 throughput on live AI Gateway traffic, in tokens per second (TPS). Visit the docs for more info.
Copy link to headingLatency24 hours
P50 time to first token (TTFT) on live AI Gateway traffic, in milliseconds. View the docs for more info.
Getting started
Call Kimi K2.6 through AI Gateway with the AI SDK generateText and streamText functions, or through the OpenAI Chat Completions, OpenAI Responses, and Anthropic Messages APIs by changing the base URL. AI Gateway authenticates the request and routes it to an available provider.
Install the AI SDK (pnpm add ai dotenv), create an API key from the API Keys page, and set it as AI_GATEWAY_API_KEY in your environment. Full setup is covered in the text generation quickstart.
import { generateText } from 'ai';import 'dotenv/config';
async function main() { const result = await generateText({ model: 'moonshotai/kimi-k2.6', prompt: 'Why is the sky blue?', });
console.log(result.text);}
main().catch(console.error);Top-level parameters
The same Kimi K2.6 request in each API format AI Gateway supports.
import { generateText } from 'ai';import 'dotenv/config';
async function main() { const result = await generateText({ model: 'moonshotai/kimi-k2.6', system: 'You are a concise technical assistant.', prompt: 'Summarize the tradeoffs between static generation and SSR.', maxOutputTokens: 1024, });
console.log(result.text);}
main().catch(console.error);Standard parameters like prompt, messages, temperature, and tools work as documented in the AI SDK docs. These are the parameters with model-specific behavior.
| Parameter | Type | Required | Description |
|---|---|---|---|
model | string | Yes | Model ID in the form creator/model, e.g. moonshotai/kimi-k2.6. AI Gateway routes the request to an available provider. |
maxOutputTokens | number | No | Hard cap on generated tokens. Kimi K2.6 supports up to 262,144 output tokens. Reasoning tokens count toward this limit. |
reasoning | 'provider-default' | 'none' | 'minimal' | 'low' | 'medium' | 'high' | 'xhigh' | No | Provider-agnostic reasoning effort, available in AI SDK 7 or later. Maps to the provider’s native reasoning configuration; reasoning settings under providerOptions take precedence when both are set. See the Reasoning section below. |
providerOptions | Record<string, JSONValue> | No | AI Gateway routing options under gateway, plus any provider-native options under the provider’s own namespace — see the table below. |
Input limits
| Input | Formats | Sources | Max count | Max size | Limits |
|---|---|---|---|---|---|
| Text | — | — | — | — | Prompt and response share the 262K-token context window |
| Image | — | URL, base64, Uint8Array | — | — | Sent as image parts in messages; counts as input tokens |
Provider options
Set AI Gateway routing options under providerOptions.gateway. For provider-specific options, pass them under the provider’s namespace as documented by the AI SDK.
Learn more in the AI SDK moonshotai provider docs.
import { generateText } from 'ai';import 'dotenv/config';
async function main() { const result = await generateText({ model: 'moonshotai/kimi-k2.6', prompt: 'Why is the sky blue?', providerOptions: { gateway: { only: ['moonshotai', 'fireworks'], }, }, });
console.log(result.text);}
main().catch(console.error);These AI Gateway routing options apply to every model. Provider-specific options pass through under the provider’s own namespace (for example providerOptions.anthropic) exactly as documented by the AI SDK.
| Parameter | Type | Required | Description |
|---|---|---|---|
providerOptions.gateway.only | string[] | No | Restrict routing to these provider slugs. Requests fail over only within the listed providers. |
providerOptions.gateway.order | string[] | No | Preferred provider order. Listed providers are tried first; unlisted providers remain available as fallbacks. |
providerOptions.gateway.sort | 'cost' | 'ttft' | 'tps' | No | Rank candidate providers by price, time to first token, or tokens per second instead of the default routing order. |
providerOptions.gateway.zeroDataRetention | boolean | No | Route only to providers with a zero-data-retention policy for this model. |
Routing across providers
AI Gateway serves the same model through multiple providers and fails over automatically. order expresses a preference while keeping every provider eligible; only is a hard allowlist — if none of the listed providers are available the request fails instead of falling back.
Options under a provider's own namespace (for example providerOptions.anthropic) are forwarded to that provider with the request. Providers ignore option namespaces that don't apply to them, so it is safe to set provider options alongside gateway routing options.
Reasoning
AI Gateway bridges reasoning across every API format. The AI SDK exposes a provider-agnostic top-level reasoning level (none, minimal, low, medium, high, or xhigh); the Chat Completions and Responses formats take the same effort under reasoning.effort; and the Anthropic Messages format uses a native thinking token budget. Whichever you send, the gateway maps it to the target model’s native configuration, converting between effort levels and token budgets as needed. Reasoning-related settings under providerOptions take full precedence over the top-level reasoning value and are never merged. Reasoning tokens typically count toward your output-token usage, though how they’re reported and billed varies by provider.
Learn more in the AI Gateway reasoning guide.
import { generateText } from 'ai';import 'dotenv/config';
async function main() { const result = await generateText({ model: 'moonshotai/kimi-k2.6', prompt: 'Explain the Monty Hall problem step by step.', reasoning: 'high', });
console.log(result.text);}
main().catch(console.error);Image input
Send images alongside text as message parts. Images count as input tokens.
import { generateText } from 'ai';import 'dotenv/config';
async function main() { const result = await generateText({ model: 'moonshotai/kimi-k2.6', messages: [ { role: 'user', content: [ { type: 'text', text: 'Describe this image.' }, { type: 'image', image: 'https://example.com/photo.jpg' }, ], }, ], });
console.log(result.text);}
main().catch(console.error);Tool calling
Expose tools the model can call. Define each tool’s inputs with a Zod schema.
import { generateText, tool } from 'ai';import { z } from 'zod';import 'dotenv/config';
async function main() { const result = await generateText({ model: 'moonshotai/kimi-k2.6', prompt: 'What is the weather in San Francisco?', tools: { getWeather: tool({ description: 'Get the current weather for a location', inputSchema: z.object({ location: z.string() }), execute: async ({ location }) => ({ location, temperatureC: 18 }), }), }, });
console.log(result.text);}
main().catch(console.error);Copy link to headingMore models by Moonshot AI
| Model |
|---|
| 1M | 0.8 s | 172 tps | $4.50/M | $22.50/M | Read$0.45/M | — | +2 |   | 07/27/2026 | |||||
| 1M | 0.5 s | 151 tps | $2.50/M+1 more | $12.75/M+1 more | Read$0.29/M | — | +2 |    +11 | 07/16/2026 | |||||
| 262K | 0.9 s | 193 tps | $1.90/M | $8/M | Read$0.38/M | — | +2 | | 06/15/2026 | |||||
| 262K | 0.7 s | 206 tps | $0.74/M+1 more | $3.50/M+1 more | Read$0.15/M | — | +2 |  +1 | 06/12/2026 | |||||
| 262K | 0.7 s | 45 tps | $0.60/M | $3/M | Read$0.10/M | — | +1 |   | 01/26/2026 | |||||
| 216K | 0.6 s | 30 tps | $0.47/M | $2/M | Read$0.14/M | — | | 11/06/2025 |
Copy link to headingAbout Kimi K2.6
Kimi K2.6, released on April 20, 2026, is the natively multimodal successor in the Kimi line. Moonshot AI positions it around three capability areas: long-horizon execution, agentic coding, and design with code.
Long-horizon coding is the headline shift from earlier K2 variants. Kimi K2.6 sustains tool-use chains across thousands of calls in a single session, with reported workloads spanning many hours of continuous execution and iterative optimization across a codebase. The model maintains task state across these extended sessions rather than losing thread after a few dozen turns.
Native vision input changes how design tasks compose. Kimi K2.6 accepts images directly, so a screenshot or mockup can drive a frontend generation step without a separate vision model in the pipeline. Moonshot AI documents output that includes structured layouts, hero sections, interactive elements, and animations, rather than syntax-level scaffolding alone. Full-stack workflows that pair frontend output with authentication, user interaction, and database operations are part of the documented scope.
Access Kimi K2.6 through AI Gateway by setting the model string to moonshotai/kimi-k2.6. AI Gateway routes across Moonshot AI, Fireworks, Novita AI, Baseten with automatic failover, and the observability layer tracks token usage and costs across the long sessions this model is built for.
Kimi K2.6 supports a context window of 262.1K tokens and completions up to 262.1K tokens per request. It's available through AI Gateway at $0.95 per million input tokens and $4 per million output tokens.
Copy link to headingWhat To Consider When Choosing a Provider
- Configuration: Kimi K2.6 runs the longest agentic sessions in the Kimi family. Plan token budgets around extended tool-use chains and verify your agent harness handles multi-hour execution windows. Vision input is native, so a separate vision model isn't required for design or screenshot tasks.
- Zero Data Retention: Zero Data Retention is available for this model. It is offered on a per-provider and model basis. See the documentation for details.
- Authentication: AI Gateway authenticates requests using an API key or OIDC token. You do not need to manage provider credentials directly.
Copy link to headingWhen to Use Kimi K2.6
Best for
- Long-horizon coding agents: Sessions that run for hours, accumulate thousands of tool calls, and iterate across a full codebase
- Vision-to-frontend pipelines: Frontend generation from screenshots, mockups, or design references without a separate vision step
- Full-stack scaffolding: Workflows spanning UI, authentication, user interaction, and database operations from one model
- Kimi K2.5 upgrade path: Teams that want stronger long-horizon execution and design output than K2.5 provides
Consider alternatives when
- Explicit reasoning traces: Kimi K2 Thinking emits chain-of-thought output for tasks that reward visible deliberation
- Short-horizon throughput: Kimi K2 Turbo runs the K2 MoE without the longer-session emphasis when tasks finish in a few turns
- Cost-sensitive deployments: Earlier K2 variants may meet your quality bar at lower cost per token
- Text-only pipelines: A text-only K2 variant is a closer fit when no vision step is needed
Copy link to headingConclusion
Kimi K2.6 extends the Kimi line into long-horizon agentic coding and design-with-code workflows with native vision input. For agents that need to run for hours across a codebase, or for pipelines that turn visual references into working frontends, it's the K2 generation built for those sessions.
Copy link to headingFrequently Asked Questions
What makes Kimi K2.6 different from Kimi K2.5?
Long-horizon execution and design output. Moonshot AI documents sustained tool-use chains across thousands of calls and frontend generation with structured layouts, hero sections, and animations. Vision input is native, so screenshots and mockups feed directly into the model without a separate vision step.
How long can Kimi K2.6 sustain a single agentic session?
Moonshot AI reports workloads spanning many hours of continuous execution with thousands of tool calls in a single session. Plan your agent harness around extended runs and budget tokens for the accumulated history.
Does Kimi K2.6 accept image inputs directly?
Yes. Kimi K2.6 is natively multimodal, so screenshots, mockups, and design references go in alongside text prompts. Confirm modality limits on https://platform.kimi.ai/docs/pricing/chat before you build a vision-heavy pipeline.
What kind of frontend code does Kimi K2.6 produce?
Moonshot AI documents output with structured layouts, hero sections, interactive elements, and animations, plus full-stack scaffolding that spans authentication, user interaction, and database operations.
How do I switch from an earlier Kimi K2 variant to kimi-k2.6?
Update the model string in your API call to
moonshotai/kimi-k2.6. Authentication, tool-calling format, and the rest of the integration stay the same.How do I use Kimi K2.6 on AI Gateway?
Use the identifier
moonshotai/kimi-k2.6with the AI SDK or any supported interface like Chat Completions, Responses, or Messages. AI Gateway routes acrossmoonshotai,fireworks,novita,basetenand handles failover automatically.Does Kimi K2.6 support zero data retention?
Yes, Zero Data Retention is available for this model. Zero Data Retention is offered on a per-provider basis. See https://vercel.com/docs/ai-gateway/capabilities/zdr for details.