MoonshotAI: Kimi K3

Name: Kimi K3 API
Brand: moonshotai
Price: 3 USD
Availability: InStock

moonshotai/kimi-k3

Date de publication: 2026-07-16

1,048,576 context · $3.00/M input tokens · $15.00/M output tokens

Kimi K3 is Moonshot AI's flagship open-weight multimodal reasoning model. It is built for complex coding, knowledge work, and long-horizon agentic workflows, with strong performance on large codebase navigation, tool use, debugging, visual reasoning, and iterative problem solving. WaveSpeed AI exposes moonshotai/kimi-k3 through an OpenAI-compatible API, so it can be used with standard OpenAI SDKs and existing chat-completions-based application flows.

Tarification

Paiement à l'usage

Aucun coût initial, payez uniquement ce que vous utilisez

Entrée$3.00 / M Tokens

Sortie$15.00 / M Tokens

Cache Read$0.30 / M Tokens

Essayer le modèle

moonshotai/kimi-k3

En ligne

Bonjour ! Je suis un assistant IA utile. Que puis-je faire pour vous ?

Prêt à utiliser ce modèle dans un coding agent local ?Setup agent

Utilisation de l'API

Utilisez les exemples de code suivants pour intégrer notre API :

import OpenAI from 'openai';

if (!process.env.WAVESPEED_API_KEY) throw new Error('Set WAVESPEED_API_KEY');
const client = new OpenAI({
  apiKey: process.env.WAVESPEED_API_KEY,
  baseURL: 'https://llm.wavespeed.ai/v1',
  timeout: 120_000,
  maxRetries: 0,
});

try {
  const response = await client.chat.completions.create({
    model: 'moonshotai/kimi-k3',
    messages: [{ role: 'user', content: 'Hello!' }],
  });
  const body = response as typeof response & { error?: { message?: string } };
  if (body.error) throw new Error(body.error.message || 'LLM generation failed');
  console.log(response.choices[0]?.message?.content ?? '');
} catch (error) {
  console.error('LLM request failed:', error);
  process.exitCode = 1;
}

import OpenAI from 'openai';

if (!process.env.WAVESPEED_API_KEY) throw new Error('Set WAVESPEED_API_KEY');
const client = new OpenAI({
  apiKey: process.env.WAVESPEED_API_KEY,
  baseURL: 'https://llm.wavespeed.ai/v1',
  timeout: 120_000,
  maxRetries: 0,
});

try {
  const response = await client.chat.completions.create({
    model: 'moonshotai/kimi-k3',
    messages: [{ role: 'user', content: 'Hello!' }],
  });
  const body = response as typeof response & { error?: { message?: string } };
  if (body.error) throw new Error(body.error.message || 'LLM generation failed');
  console.log(response.choices[0]?.message?.content ?? '');
} catch (error) {
  console.error('LLM request failed:', error);
  process.exitCode = 1;
}

Introduction au modèle

Kimi K3 is a 2.8T parameter open-weight multimodal reasoning model from Moonshot AI. It is suited for complex coding, knowledge work, and long-horizon agentic workflows, and is particularly strong at...

This model is available through the WaveSpeed AI OpenAI-compatible API for chat completions and compatible application workflows.

Why It Looks Great

text+image+video->text architecture for Text, Image, Video to Text workloads
1048576 context window for long prompts, document analysis, and multi-turn workflows
Vision input support for image understanding and multimodal tasks
Function calling and tool-use support for agentic application workflows
Structured output support for JSON responses and schema-constrained generation
Reasoning controls available through supported parameters

Key Features

Context Window: 1048576 tokens
Max Input: 104858 tokens
Max Output: 943718 tokens
Vision: Supported
Function Calling: Supported
Structured Outputs: Supported
Image Generation: Not listed
Audio Input: Not listed
Supported Parameters: frequency_penalty, include_reasoning, logit_bias, logprobs, max_completion_tokens, max_tokens, min_p, parallel_tool_calls, presence_penalty, reasoning, reasoning_effort, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_logprobs, top_p

Specifications

Specification	Value
Provider	moonshot
Model Type	Chat Completions model
Architecture	text+image+video->text
Context Window	1048576 tokens
Max Input	104858 tokens
Max Output	943718 tokens
Input	Text, Image, Video
Output	Text
Vision	Supported
Function Calling	Supported
Structured Outputs	Supported
Release Date	July 16, 2026

How to Use

Write your prompt - describe the task, provide context, and specify the desired output format.
Submit - the model processes your request and returns the response.

API Integration

Base URL: https://llm.wavespeed.ai/v1 API Endpoint: chat/completions Model ID: moonshotai/kimi-k3

API Usage

Python SDK

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://llm.wavespeed.ai/v1"
)

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

print(response.choices[0].message.content)

cURL

curl https://llm.wavespeed.ai/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -d '{
    "model": "moonshotai/kimi-k3",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Notes

Model: moonshotai/kimi-k3
Provider: moonshot

Infos

Fournisseurmoonshot

Typellm

Fonctionnalités prises en charge

Entrée

TexteImage

Sortie

Texte

Contexte1,048,576

Sortie max-

Vision✓ Pris en charge

Function Calling✓ Pris en charge

Guide d'accès API

Base URLhttps://llm.wavespeed.ai/v1

API Endpointchat/completions

ID du modèlemoonshotai/kimi-k3

Kimi K3 API

moonshotai/kimi-k3

Kimi K3 is Moonshot AI's flagship open-weight multimodal reasoning model. It is built for complex coding, knowledge work, and long-horizon agentic workflows, with strong performance on large codebase navigation, tool use, debugging, visual reasoning, and iterative problem solving. WaveSpeed AI exposes `moonshotai/kimi-k3` through an OpenAI-compatible API, so it can be used with standard OpenAI SDKs and existing chat-completions-based application flows.

Questions fréquentes sur Kimi K3

Combien coûte l'API Kimi K3 ?+

Tarification sur WaveSpeedAI : $3.00 par million de tokens d'entrée et $15.00 par million de tokens de sortie. Le prompt caching et le traitement par batch sont facturés séparément et réduisent le coût effectif sur les charges longues et répétitives.

Quelle est la fenêtre de contexte de Kimi K3 ?+

Kimi K3 prend en charge jusqu'à 1049K tokens de contexte et jusqu'à — tokens de sortie par requête.

Kimi K3 est-il compatible avec OpenAI ?+

WaveSpeedAI expose Kimi K3 à https://llm.wavespeed.ai/v1 via l’interface Chat Completions compatible OpenAI. La plupart des clients du SDK OpenAI fonctionnent en changeant l’URL de base et la clé API ; les champs facultatifs dépendent du modèle.

Comment démarrer avec Kimi K3 ?+

Connectez-vous à WaveSpeedAI, créez une clé API dans Access Keys, puis envoyez une requête à https://llm.wavespeed.ai/v1/chat/completions avec l’id du modèle affiché ci-dessus. Consultez le catalogue actuel pour la disponibilité, les capacités et les tarifs.

Tarification

Essayer le modèle

Utilisation de l'API

Introduction au modèle