Hormouz AIHormouzSign in

MoonshotAI: Kimi K3: AI model

Kimi K3 is a 2.8-trillion-parameter open-weight multimodal reasoning model from Moonshot AI with a 1M-token context window. Reachable from inside Iran through Hormouz.

Context window:
1,048,576 tokens
Input:
$0.73/1M tokens
Output:
$10/1M tokens
Type:
text, image, video
  • Vision
  • Tools
  • Streaming
  • Long context

Thinking levels: max · high · low

Kimi K3 is Moonshot AI's open-weight multimodal reasoning model at 2.8 trillion parameters. It is built for complex coding, knowledge work and long agent workflows, and is especially good at finding its way around large repositories, using tools, debugging and iterating against images, logs, tests and runtime feedback. Its KDA and Attention Residuals design keeps that scale efficient.

It accepts text, image and video input and returns text, and supports tool calling, structured JSON output and streaming. Reasoning can be switched on when a problem needs more thought and left off when speed matters. Kimi K3 is commonly used for large-repository coding, debugging against logs and tests, long agent workflows and knowledge work. Hormouz serves it behind an OpenAI-compatible API: point your base URL at Hormouz, drop in a Hormouz key, and existing OpenAI SDK code runs without changes.

Pricing is usage-based: $0.67 per million input tokens and $14 per million output tokens. There is no fixed monthly fee, and you pay only for the tokens you use. Hormouz provides stable access to the model from inside Iran, with no foreign credit card or circumvention tooling required.

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...

Quickstart

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

Specs

VendorMoonshot AI
TypeText, Image, Video
Context window1,048,576 tokens
Pricing$0.73/M in · $10.00/M out
APIChat Completions
CapabilitiesStreaming · Vision · Tools

Benchmarks

30.1
Intelligence Index
Better than 57% of models we compared
72
Coding Index
Better than 87% of models we compared
29.3
Agentic Index
Better than 66% of models we compared

Reasoning

GPQA Diamond
Graduate-level scientific reasoning
84.2%
HLE
Humanity's Last Exam
25%
AA-LCR
Long-context reasoning
79.3%
CritPt
Research-level physics reasoning
3.1%
AA Omniscience
Broad factual knowledge accuracy
45.8%
Non-Hallucination
Rate of avoiding fabricated answers
22.9%

Coding

SciCode
Python for scientific computing
52.7%

Agentic

GDPval
Economically valuable tasks
31.4%

Source: Artificial Analysis

Agentic Game Dev
Top 12%
1,252Elo
Win rate: 55%
Android Native
Top 14%
1,262Elo
Win rate: 55.4%
Full-Stack
Top 5%
1,313Elo
Win rate: 61.8%
Godot Game Dev
Top 39%
1,223Elo
Win rate: 48.2%
HTML Slides
Top 9%
1,256Elo
Win rate: 59.6%
Mobile Apps
Top 7%
1,266Elo
Win rate: 56.9%
Python Slides
Top 11%
1,273Elo
Win rate: 59.2%
Web Apps
Top 12%
1,285Elo
Win rate: 59.1%
3D
Top 3%
1,402Elo
Win rate: 68%Avg time: 635s
Code
Top 1%
1,372Elo
Win rate: 63.4%Avg time: 714s
Data Viz
Top 2%
1,358Elo
Win rate: 64%Avg time: 643s
Game Dev
Top 2%
1,379Elo
Win rate: 61.8%Avg time: 848s
SVG
Top 7%
1,299Elo
Win rate: 59.4%Avg time: 429s
UI Component
Top 2%
1,362Elo
Win rate: 62.3%Avg time: 695s
Website
Top 2%
1,344Elo
Win rate: 59.9%Avg time: 773s

Metrics sourced from Design Arena

Code example

cURL
curl https://api.hormouz.net/v1/chat/completions \
  -H "Authorization: Bearer ng-YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "moonshotai/kimi-k3",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'
Python
from openai import OpenAI
client = OpenAI(api_key="ng-YOUR_API_KEY", base_url="https://api.hormouz.net/v1")
response = client.chat.completions.create(
    model="moonshotai/kimi-k3",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)
JavaScript
import OpenAI from "openai";
const client = new OpenAI({ apiKey: "ng-YOUR_API_KEY", baseURL: "https://api.hormouz.net/v1" });
const response = await client.chat.completions.create({
  model: "moonshotai/kimi-k3",
  messages: [{ role: "user", content: "Hello!" }],
});
console.log(response.choices[0].message.content);

Buy the MoonshotAI: Kimi K3 API, paid in rial

  • Buy credit with an Iranian bank card; no foreign card or account needed.
  • One Hormouz key reaches MoonshotAI: Kimi K3 and 600+ other models, and you pay only for what you use.
  • No VPN and no risk of a banned account; just change the base URL to api.hormouz.net/v1.
Create an account

Use via Hormouz

Connect to this model and 600+ others with one API key, then see usage down to each request.

Request company access

FAQ

How much does Kimi K3 cost?
Pricing is usage-based: $0.67 per million input tokens and $14 per million output tokens. There is no fixed monthly fee, and you pay only for the tokens you use.
How do I access Kimi K3 from inside Iran?
Create an account on Hormouz and generate an API key. Hormouz provides a stable path to the model from inside Iran, with no foreign credit card or circumvention tooling required.
Is Kimi K3 compatible with OpenAI code?
Yes. Hormouz exposes an OpenAI-compatible API. Point your base URL at Hormouz and swap in your Hormouz key. Your existing SDKs and code run without a rewrite.
What context window and capabilities does Kimi K3 have?
It has a 1M-token context window, accepts text, image and video input, and supports tool calling, structured JSON output and streaming. Reasoning can be switched on when a problem needs more thought and left off when speed matters.
What is Kimi K3 good for?
Typical uses are large-repository coding, debugging against logs and tests, long agent workflows and knowledge work.

Updated 2026-10-13