Skip to content
03 Model garden Serve

Kimi K3

Instantly via the EU router or as a dedicated GPU deployment. Data stays in Europe.

curl
$ curl https://hostyourai.com/api/v1/chat/completions \
  -H "Authorization: Bearer hyai-..." \
  -H "Content-Type: application/json" \
  -d '{
    "model": "moonshotai/Kimi-K3",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Call it now

Drop-in replacement for OpenAI: change only the base URL and API key. The Anthropic format (/v1/messages) is supported too.

Serve Model garden
moonshotai/Kimi-K3

text->text · moonshotai · sovereign EU

Runs on EU infrastructure operated by European companies, from your request to the GPU. See the full chain

Kimi K3 is an open-weight, native multimodal agentic model and our most capable model to date. It is a 2.8T-parameter model built on Kimi Delta Attention (KDA) and Attention Residuals (AttnRes), with native vision capabilities and a 1-million-token context window. It is the world...

Router
Router · warm
Dedicated
Dedicated · available
Input (per 1M tokens)
€3.45
Output (per 1M tokens)
€17.25
1,049KContext window
1680GBMinimum VRAM
POST /api/v1/chat/completions 200 OK

Specifications

Context window1,048,576 tokens
Minimum VRAM1680 GB
ArchitectureKimiK3ForConditionalGeneration (vLLM)
Licenseopen-weights
Modalitytext->text
ReleasedJune 2026
Publishermoonshotai ↗

Pricing

Shared router · per token

€3.45Input (per 1M tokens)
€17.25Output (per 1M tokens)

€0.86 per million cached input tokens

Dedicated GPU · per hour

from €83,40 per hourYour own vLLM instance on European cloud (1680 GB VRAM), billed per minute.

Shared EU router, you pay only per token. Dedicated GPU deployments are billed per minute, see pricing.

Bench index

Bench is our quality index for every model, 0 to 100. It rests on public measurements (Epoch AI with about 40 benchmarks, LiveBench, LMArena and more) and is refreshed every morning. Where nobody measures, it shows an estimate with a range.

85.8 / 100 · Measured

Measured by Epoch AI across 22 benchmarks, with a range.

Range 83.3 to 88.5 · Rank 26 of 258 at Epoch AI

Sources: Epoch AI ↗ · LiveBench ↗ · LMArena ↗ · Refreshed daily, last on 22-11-2026

LiveBench · 79.2 · Rank 16 of 70

Reasoning90.7
Coding81.5
Agentic coding62.2
Mathematics84.4
Data analysis78.7
Language85.5
Instruction following71.4
Every benchmark (22)
OTIS Mock AIME 2024-202597.2
Surface Evolver Bench95.0
GPQA diamond94.6
ARC-AGI94.5
ProofBench87.0
DTBench85.3
WeirdML82.6
FrontierMath-Tiers-1-3-v2-Private76.0
LMCA62.0
ARC-AGI-260.4
MirrorCode54.4
SimpleBench52.8
SimpleQA Verified50.6
APEX-Agents47.8
FrontierCode44.2
FrontierMath-Tier-4-v2-Private39.0
Chess Puzzles33.7
PostTrainBench31.5
FrontierSWE25.9
Mystery Game Puzzles18.5
WeirdML v317.1
Furniture Assembly6.0
LiveBench · All tasks
Reasoning
theory of mind82.7
zebra puzzle100.0
spatial100.0
logic with navigation80.0
Coding
code generation80.3
code completion82.6
Agentic coding
javascript68.2
typescript53.3
python65.0
Mathematics
AMPS Hard97.0
integrals with game54.0
math comp95.1
olympiad91.7
Data analysis
consecutive events89.8
tablejoin48.4
tablereformat98.0
Language
connections100.0
plot unscrambling72.6
typos84.0
Instruction following
paraphrase74.4
simplify66.4
story generation75.4
summarize69.4
Questions

Frequently asked questions

Can I run Kimi K3 in the EU?

Yes. HostYourAI runs Kimi K3 on GPUs in European datacenters via vLLM. Prompts and outputs never leave the EU and there is no US cloud provider in the chain.

Is hosting Kimi K3 GDPR-compliant?

Yes. All processing happens inside the EU, a Data Processing Agreement (DPA) is available and the subprocessor list is public. Open-source weights also mean: no training on your data.

How much does Kimi K3 cost?

Via the shared EU router you pay €3.45 per million input tokens and €17.25 per million output tokens, with no fixed costs. For high volume or isolation you can also run Kimi K3 on a dedicated GPU instance, billed per minute.

Is the API OpenAI-compatible?

Yes. You use the standard OpenAI SDKs with a custom base URL (https://hostyourai.com/api/v1). The Anthropic Messages API is supported as a drop-in as well.

More models from Moonshotai

The platform

One platform, four layers

Start with one layer and add the others when the work asks for it. They share one key, one balance and one Activity view.

Code

Your coding agent on open models in Europe, set up with one command.

Open

Control

Keys, budgets and guardrails per team, and personal data stopped before any model sees it.

Open

Serve

Shared per token, or running for you alone, from European data centres.

Open

EU compute

Private GPUs in Europe, H100 to B300, sized for response time, generation speed and concurrent users.

Open

Try Kimi K3 for free

Creating an account takes a minute. Test Kimi K3 straight away in the playground.

  • Jantine
  • Hans
  • Sander
  • Jos
  • Daan
  • Kyra
  • Margriet
  • Stan
The team behind HostYourAI