Moonshotai logo

Kimi K3 – Chat and Inference API hosted in Europe

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...

Input Price

$4.00/1M tokens

Output Price

$18.00/1M tokens

Intelligence

46.5

Coding

N/A

For prompts longer than 131,072 tokens, every token type (input, output, cache) is billed at 1.25× the price.Applies to existing customers from Nov 18, 2026; annual plans from their first renewal after that date

Specifications

Technical details and pricing.

ProviderMoonshotai
Context Window1,048,576 tokens
Release DateOct 23, 2026
ModalitiesText, Image → Text
CapabilitiesFunction Calling, Structured Outputs, JSON Mode, Vision
AvailabilityChat, Inference API

EU-hosted inference API

Power your AI projects with open-source models.

Drop-in OpenAI-compatible API. No data leaves Europe.

MiniMax

MiniMax M3

$0.50 / $1.90

per M tokens

Z.ai

GLM 5.3 Flash

$0.20 / $0.60

per M tokens

MoonshotAI

Kimi K3

$4.00 / $18.00

per M tokens

DeepSeek

DeepSeek V4.1 Flash

$0.40 / $1.42

per M tokens

Frequently Asked Questions

What is Kimi K3 good for?

Use Kimi K3 for everyday tasks like writing, summarizing, brainstorming, and getting clear explanations.

How much does Kimi K3 cost?

Pricing is based on usage. Current rates are $4.00/1M tokens for input and $18.00/1M tokens for output. For prompts longer than 131,072 tokens, every token type (input, output, cache) is billed at 1.25× the price. (Applies to existing customers from Nov 18, 2026; annual plans from their first renewal after that date.)

Can I use Kimi K3 in LLMBase Chat?

Yes. Use the chat action on this page to open Kimi K3 in LLMBase Chat. Access still follows the current plan and model tier.

Does Kimi K3 support images or audio?

Kimi K3 can understand images.

Where does Kimi K3 run?

Kimi K3 is hosted by LLMBase in EU data centers. LLMBase is operated by a German company, and a DPA is available for business customers.

Benchmarks and pricing use Artificial Analysis where available. Catalog specs are used as a fallback.