Chinese model APIs on one OpenAI-compatible key. Browse models
Models / Moonshot AI
Kimi logo for the Kimi K2 API model page
Kimi API

Kimi K2 API

Access Kimi family models for long-context reading, agent planning, coding and multimodal-ready workflows through one compatible API.

Integrating · catalogued for evaluation while the dedicated route is being wired

Model ID

kimi-k2

Base URL

https://thesmartoken.com/v1

Input / 1M

$0.6

Output / 1M

$2.5

reasoningcodingtool callinglong context256K context
Long-context workspaces
Agentic coding flows
OpenAI SDK migration
Budget controls per key
curl --fail-with-body -sS "https://thesmartoken.com/v1/chat/completions" \
  -H "Authorization: Bearer $SMTOKEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "kimi-k2",
    "messages": [{"role":"user","content":"Reply with exactly OK."}],
    "max_tokens": 512
  }'

Kimi long-context route

OpenAI-compatible model gateway

Integrating

Client

</>

Your App

SDK, backend, agent or workflow.

SmarToken

Unified API Gateway

kimi-k2

AuthBudgetRoute
SSE stream

Provider

Kimi K2

Moonshot AI route through SmarToken.

01

Your app

Send a standard Chat Completions request with your SmarToken key.

02

SmarToken gateway

Validate the key, model scope, daily budget and monthly budget.

03

Model route pool

Choose an enabled route by priority and weight. No automatic fallback or cross-provider failover is promised.

04

Moonshot AI

Call kimi-k2, stream the response and record usage.

Example JSON response

200 OK
{
  "model": "kimi-k2",
  "choices": [{
    "message": {
      "role": "assistant",
      "content": "Kimi is useful when your app needs to read longer documents, repositories and research material before answering."
    },
    "finish_reason": "stop"
  }],
  "usage": { "prompt_tokens": 10, "completion_tokens": 12, "total_tokens": 22 }
}

Fast Deployment

Generate a key and run the Playground in minutes.

Global Access

Use one endpoint for configured China-first routes.

Unified Billing

Track USD token usage and wallet debits together.

Easy Migration

Keep OpenAI SDK shape, change baseURL and model.

Developer Friendly

Copy-ready cURL, Python and Node.js examples.

Model fit

Why use Kimi?

A focused page for evaluating this model as an API route, not just reading a catalog row.

Supplier capabilities described here do not mean every feature is supported by this gateway. Current support is text Chat Completions; tools, structured output and multimodal requests are rejected. Review the gateway contract.

Long context

A strong page fit for document Q&A, repository analysis and multi-step planning.

Agent workflows

Good for teams prototyping research agents and coding agents that need memory over a larger prompt.

Developer migration

Works behind the same Chat Completions interface used by the rest of SmarToken.

Route fit

Kimi route-fit checks

Treat kimi-k2 as a route to test against a workload, not just a catalog name.

When Kimi is a strong fit

Kimi K2 is strongest for Long context, agents, research. Validate it with real prompts before making it a default route.

When to compare alternatives

If your workload depends on long context or low cost, compare alternative catalog models with the same prompts.

Pre-production check

Long prompts still need careful chunking and cost monitoring.

Kimi specifics

Long-context notes that only apply to Kimi

Kimi earns a route when input length dominates the workload. These notes cover where the long-context advantage is real and where a shorter, cheaper route still wins.

Document-heavy workflows

Contracts, research reports, meeting archives and multi-file code reading are the natural fit. Send the document set plus a narrow question, and ask for cited spans so reviewers can verify answers quickly.

Long context is not free

Input tokens scale cost and latency even when the window fits. For repeated questions over the same corpus, retrieval with smaller prompts often beats resending full documents. Compare cost per useful answer, not context size.

When Kimi is not the pick

For pure coding value compare DeepSeek first, and for multilingual structured output compare Qwen. Kimi should win the route because prompts are long, not because the model name is familiar.

Why SmarToken

Why not direct vendor accounts?

Direct accounts can work once region, billing, model IDs and credentials are settled. SmarToken is built for faster overseas evaluation and safer early production.

Unified key

One API key works across the Chinese model catalog.

English docs

Model IDs, SDK samples and error semantics are written for overseas teams.

Budget control

Daily, monthly and model-family limits keep experiments predictable.

Route visibility

Usage logs connect model, API key, token estimate, latency and wallet debit.

Code sample

Copy-ready API examples

Open in Playground

curl

curl --fail-with-body -sS "https://thesmartoken.com/v1/chat/completions" \
  -H "Authorization: Bearer $SMTOKEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "kimi-k2",
    "messages": [{"role":"user","content":"Reply with exactly OK."}],
    "max_tokens": 512
  }'

python

import os
from openai import OpenAI

client = OpenAI(api_key=os.environ["SMTOKEN_API_KEY"],
                base_url="https://thesmartoken.com/v1", max_retries=0)
response = client.chat.completions.create(
    model="kimi-k2",
    messages=[{"role": "user", "content": "Reply with exactly OK."}],
    max_tokens=512,
)
print(response.choices[0].message.content)
print(response.usage)

Node.js (JavaScript)

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.SMTOKEN_API_KEY,
  baseURL: "https://thesmartoken.com/v1", maxRetries: 0,
});
const response = await client.chat.completions.create({
  model: "kimi-k2",
  messages: [{ role: "user", content: "Reply with exactly OK." }],
  max_tokens: 512,
});
console.log(response.choices[0].message.content);
console.log(response.usage);

Compare nearby Chinese AI models

Live routes plus a few nearby catalog peers. Browse the full catalog.

ModelAPI IDContextInput / 1MOutput / 1M
Kimi K2kimi-k2256K$0.6$2.5
DeepSeek V4.1 Flashdeepseek-v4.1-flash1M$0.3$1.2
DeepSeek V4.1 Flash (Vendor Direct)deepseek-tokenhub-flash1M$0.3$1.2
DeepSeek V4 Pro (TokenHub)deepseek-tokenhub-pro1M$1.74$3.48
Kimi K3.1 Prokimi-k3.1-pro1M$2$10
Kimi K2.7 Codekimi-k2.7-code256K$0.95$4
DeepSeek Flashdeepseek-chat1M$0.3$1.2
DeepSeek V4 Prodeepseek-v4-pro1M$1.32$3.96
Kimi K3kimi-k31M$3$15
GLM-5.2glm-5.21M$1.4$4.4
GLM-5.3glm-5.31M$1.4$4.4
MiniMax M2.7minimax-m2200K$0.3$1.2
Kimi K2.6kimi-k2.6262K$0.95$4
Qwen Plusqwen-plus128K$1.2$4.8
Qwen3.7 Plusqwen3.7-plus1M$0.276$1.101
GLM-4 Flashglm-4-flash128K$0.8$3.2

How do I call Kimi with the OpenAI SDK?

Use baseURL https://thesmartoken.com/v1, authenticate with a SmarToken key, and send model: kimi-k2.

When should I choose Kimi over DeepSeek or Qwen?

Choose Kimi when long-context reading, planning and document-heavy workflows matter more than the lowest per-token price.

Use cases

Popular use cases

Start with a clear workload, then compare routes in Playground before moving traffic into production.

Document Q&A

Summarize and query long reports, contracts and technical references.

Codebase reading

Use longer context for repository maps, migration notes and issue triage.

Research agent

Plan multi-step answers from larger source packets.

Customer analysis

Condense transcripts, tickets and user feedback into action lists.

Migration

From your current AI gateway

The integration path stays familiar: same Chat Completions shape, one new baseURL and a China-first model ID.

Switch path

From OpenAI

Keep the SDK. Change baseURL to SmarToken and switch the model ID.

Switch path

From OpenRouter

Move from broad routing into a focused Chinese-model console with budgets.

Switch path

From Together.ai

Keep Chat Completions while adding China-first model pages and route control.

FAQ

Questions developers ask

How do I call Kimi with the OpenAI SDK?

Use the standard OpenAI SDK, set baseURL to https://thesmartoken.com/v1, authenticate with your SmarToken key, and pass model: kimi-k2.

What is Kimi best for?

Kimi is a good candidate for long-context reading, research agents and codebase analysis. Test it in the Playground before routing production traffic.

How does Kimi compare with DeepSeek and Qwen?

Use the comparison table as a starting point, then run your own prompts because coding, reasoning, latency and cost can vary by upstream route.

Is this API compatible with the OpenAI SDK?

Yes. Use your SmarToken key, set baseURL to https://thesmartoken.com/v1, and pass the model ID shown on this page.

How is billing calculated?

Input and output tokens are priced separately in USD per 1M tokens. Final usage is recorded after the provider returns a response or a stream finishes.

Where do I get an API key?

Create an account, open the console, generate an API key and copy the cURL, Python or Node.js example from Quickstart.

Can I restrict one key to this model family?

Yes. Console API keys can be limited by allowed model family plus daily and monthly USD budgets.

Start now

Start building with Chinese AI models in minutes

One API key works across the Chinese model catalog.