Home → Pricing → kimi-k3

Kimi K3 API pricing

Name: kimi-k3 API
Brand: APICLAN
Price: 0.59 USD
Availability: InStock

kimi-k3

The current Kimi flagship. Roughly three times K2.7 on input and nearly four times on output, so it earns its place on the final reasoning step rather than on volume.

Kimi You pay 20% of list price OpenAI-compatible Pay as you go

List price · input

$2.95

per 1M tokens · output $14.74

You pay · input

$0.590

per 1M tokens · output $2.95

You save 80% on every call. Same model, same context window, same speed — the only thing that changes is your bill.

Price breakdown

Per 1M tokens	List price	APICLAN	You save
Input	$2.95	$0.590	80%
Output	$14.74	$2.95	80%

Cached input is billed at a fraction of the input rate, so agent tools that resend the same context — Codex, Claude Code, Cline — cost considerably less in practice than the raw token count suggests.

Work out your own cost

Put in the token counts from a typical request. Both columns update as you type.

Input tokens

Output tokens

Number of requests

At list price

—

At APICLAN

—

What a real session costs

A coding session through Codex on this platform used 90,459 tokens across five requests — roughly seventeen minutes of back-and-forth over a codebase. At list price that is $0.178. The customer paid $0.036.

That is real billing data, not an estimate. Heavy context reuse is exactly where the saving compounds: the same work, one fifth of the invoice.

How to call kimi-k3

The API is OpenAI-compatible. Point your base URL at APICLAN and pass the model name — nothing else in your code changes.

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_KEY",
    base_url="https://apiclan.us/v1",
)

resp = client.chat.completions.create(
    model="kimi-k3",
    messages=[{"role": "user", "content": "Hello"}],
)

Using Claude Code or the Anthropic SDK? Those clients append /v1/messages themselves, so their base URL is https://apiclan.us with no /v1. Full setup for every client is in the Quickstart.

Other Kimi models

Model	Input / 1M	Output / 1M

Prices above are what you pay, not list price. Cheaper models in the same family often handle routine work at a fraction of the cost — worth testing before defaulting to the largest one.

Start using it

No subscription, no monthly minimum, no sales call. Top up with USDT and spend what you use — 1 USDT gives you 2 credits of API balance.

Read the 30-second quickstart

Pricing verified 2026-10-03 · from the published list price. List prices change; this page is regenerated automatically from live billing data.

Language: English · 简体中文 · 繁體中文 · 日本語 · 한국어 · Deutsch · Français · Español · Türkçe · Português · Italiano · Bahasa Indonesia · Tiếng Việt · Bahasa Melayu · Polski · Čeština · فارسی · Русский