Are the Kimi K3 weights open and can I run it locally?

Yes, Moonshot AI has released the full model weights publicly, so you can download and self-host them. At 2.8T total parameters, however, you will need substantial hardware; most people will use it via API rather than running it locally.

What does Kimi K3 cost and what is its context window?

Via the API it is priced at $0.73 per million input tokens and $10.00 per million output tokens, with a context window of 1,048,576 tokens — just over one million.

How does Kimi K3 differ from other frontier reasoning models?

It is the first openly released model in the 3T-parameter class, uses a novel KDA and Attention Residuals architecture, and is designed specifically for long-horizon agentic workflows with native vision — rather than being a general-purpose chat model with reasoning bolted on.

▲

23.9%

this week

Kimi K3

9095°

#llm #moonshotai #vision #video #reasoning #coding

Run this model

Offers / Ads

OpenRouter Terminal / API

Pay per token

Chat with this model in the browser or call it from your code with one API key.

MoonshotAI | 1,048,576 token context | $0.30/M input, $11.94/M output. 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 navigating large repositories, using tools, debugging, and iterating against images, logs, tests, and runtime feedback. Its architecture uses KDA and Attention Residuals for computational efficiency.

Guides for this model

Open weights vs open source vs API-only: what you are actually allowed to do

How to choose an AI model in October 2026: the closed source news, then the open weights worth downloading

Best local LLM by use case (October 2026): the right open model for every job

How much VRAM do I actually need for local AI?

How to Install Ollama and Run a Local AI (Windows, Mac, Linux)

What is a GGUF file? Quantized models explained (Q4, Q8, fp8, NF4)

All guides

Captain's notes

Written by Captain, the house editor

Notes from 2026-10-13

Kimi K3 is a 2.8-trillion-parameter open-weight multimodal reasoning model from Moonshot AI, built on a novel KDA and Attention Residuals architecture with 104B parameters activated per token. It handles text, images, and video natively across a one-million-token context window, and is aimed squarely at long-horizon agentic work: multi-step coding sessions, large-repo navigation, and tool-heavy research pipelines. If you need a frontier-class open model for serious engineering or knowledge work, this is the one to evaluate.

Known for

Sustains long agentic coding sessions across massive repositories with minimal oversight.
Native multimodal: understands text, images, and video within a single model.
One-million-token context window for deep document and codebase work.
2.8T-parameter MoE architecture activating 104B per token for frontier-level reasoning.
Open weights released by Moonshot AI for research and deployment.

Watch out for

Output is priced at $10.00 per million tokens — costs climb fast in long agentic loops.
At 2.8T total parameters, local deployment demands serious hardware; most users will hit the API.
Weights are released under the Kimi K3 License, not a standard open-source licence — check its terms before commercial use.

Best for

Prompts to try

Debug a failing test suite

I have a Python repository where the integration tests are failing after a recent refactor. Here is the error log and the relevant files. Identify the root cause, trace it through the codebase, and propose a minimal fix with an explanation of why the original change broke the tests.

Reason through a system design

Design a rate-limiting service for a public API that must handle 50,000 requests per second, support per-user and per-endpoint limits, degrade gracefully under Redis failure, and add no more than 2ms of p99 latency. Walk through your architecture decisions and the trade-offs you are making at each step.

Analyse a screenshot for bugs

I am attaching a screenshot of my web app's dashboard. The chart in the top-right panel is rendering incorrectly — the bars are overlapping and the axis labels are cut off. Diagnose what CSS or charting-library settings are likely causing this and give me the corrected configuration.

Compare with

Kimi Latest
API-only model
1,048,576 token context · $0.29/M input, $24.00/M output
Kimi K2 Thinking
Open-weights model
262,144 token context · $0.60/M input, $2.50/M output
Kimi K2.7 Code
Open-weights model
262,144 token context · $0.67/M input, $3.35/M output
Kimi K2.5
Open-weights model
262,144 token context · $0.50/M input, $2.50/M output

More text models from moonshotai

Questions people ask

Are the Kimi K3 weights open and can I run it locally?: Yes, Moonshot AI has released the full model weights publicly, so you can download and self-host them. At 2.8T total parameters, however, you will need substantial hardware; most people will use it via API rather than running it locally.
What does Kimi K3 cost and what is its context window?: Via the API it is priced at $0.73 per million input tokens and $10.00 per million output tokens, with a context window of 1,048,576 tokens — just over one million.
How does Kimi K3 differ from other frontier reasoning models?: It is the first openly released model in the 3T-parameter class, uses a novel KDA and Attention Residuals architecture, and is designed specifically for long-horizon agentic workflows with native vision — rather than being a general-purpose chat model with reasoning bolted on.

Captain's verdict
The most capable open-weight model available right now, built for agents and serious engineering work — but the output price will punish long runs.

Model card on Hugging Face

Captain writes these notes from the model's own description and model card. Prices and context sizes are the provider's at the time of writing and can change.

Ranking badge

Updated Oct 22, 2026

Made this model? Show this badge on your page.

Paste it into your Hugging Face card or Civitai description.

Change colors / personalize / get code

Also ranked: #16 Hottest Text / LLM model · #16 Hottest Text / LLM text model

Not this creator?

Get your own badge

Kimi K3 Popularity Trend

Heat score since 16 Jul 2026

Heat score

9,095°

This week

▲

23.9%

9,067° then, 9,095° now

This month

▲

189%

8,954° then, 9,095° now

Past year

n/a

Wasn't released/tracked

7,000°

Floor

The scale runs from 0° to 9,999°. Everything below 7,000° is kept for models no source reports on yet, which carry a stand-in number there so lists stay complete. 7,000° is where measured data begins: a model with real data but almost no downloads sits at the floor, and the measurements fill the degrees above it. The floor says nothing about the model; only the degrees above it do, and they are what separates one model from another.

+1,767°

OpenRouter tokens

7,787,431,365,488 tokens this model processed through OpenRouter in the last 30 days, prompts and answers together. Every 10,000,000 tokens count as one download in the download term, 778,743 in all; the last day's tokens join the adoption rate. Source: OpenRouter (openrouter.ai/rankings).

+328°

Adoption rate

About 23,609 downloads a day, measured between two readings of the sources, never guessed from a lifetime average. Speed weighs a quarter of the download term, so a fresh surge shows without letting a one-day spike outrank years of use.

Show detailed breakdown

Every number on this site can be checked. This model's heat score is 9,095°, and here it is built one ingredient at a time, with the running total after each: the floor, the degrees each source adds, the bonus for the kind of model and what the age of its weights takes away. The colors are the bands of the chart above. The glossary at the end defines every term.

1
Start at the floor.
Why 7,000° and not zero: the scale runs from 0° to 9,999°, and the part below 7,000° is reserved for models no source reports on yet, which carry a stand-in number there so the lists stay complete and never outrank a model with data. 7,000° is where measured data begins. A model with real data but almost no downloads sits here; everything that follows is measurement, and the room above the floor holds ten tenfold steps in downloads at 300° each. The floor is the same for every model, so it decides nothing; the degrees above it do.
7,000°
running total
2
Add the downloads.
778,743 download-equivalents in all, from OpenRouter tokens. Ten times the downloads adds the same 300°, so size matters but never runs away with the score.
+1,767°
8,767°
running total
3
Add the adoption rate.
About 23,609 downloads a day right now, measured between two readings, at a quarter of the weight.
+328°
9,095°
running total
=
The heat score.
Where this model stands on every list of this site. Nothing else adjusts it.
9,095°

Glossary: the terms used here

Heat score: One number for how much a model is used and wanted right now, from 0° to 9,999°. Every list on this site that says hottest is sorted by it and nothing else.
Floor: The 7,000° every model with real data starts from. The scale runs from 0° to 9,999°; the part below the floor is reserved for stand-ins (models without data), so that a model with data always ranks above one without. The floor is the same for everyone and decides nothing; the degrees above it are the measurement.
Download-equivalents: Everything the score treats as a download, added up: downloads themselves, each torrent seeder as 1,000, each AI Horde image and each run on a generation site as one.
Log scale: Counting by tens, not by ones: going from 1,000 to 10,000 downloads adds the same degrees as going from 10,000 to 100,000. It keeps giants from crowding out everything else.
Likes: Thumbs up and likes from every source, worth half a download each on the log scale.
Adoption rate: How many downloads a day a model gets right now, from the difference between two readings of a source. A quarter of the weight of the download term.
Measured and estimated: Measured means two readings of a source exist and the difference is real. Estimated means only today's total is known and the past is spread over the model's life; the site says which one it is showing.
Seeder: Someone who keeps a copy of the weights online for others to download over a torrent, or a hosted mirror doing the same. Each one counts as 1,000 downloads.
AI Horde: A volunteer network that generates images for free on donated computers. The number of images it has made with a model is public, and each counts as one download.
Generation sites: Sites that host their own copies of a model and let people generate with it, such as PixAI, SeaArt and Yodayo. Their downloads count like Civitai's and each generation run as one download.
OpenRouter tokens: A text model served through an API, such as Claude or GPT, has no weights to download, so its usage is measured in tokens: the pieces of text it read and wrote through OpenRouter, a gateway many apps use, over the last 30 days. Every 10,000,000 tokens count as one download. Source: OpenRouter (openrouter.ai/rankings), CC BY 4.0.
Full model (checkpoint): The weights themselves, the file that generates on its own. It gets a bonus because add-ons depend on it.
LoRA and embedding: Small add-on files that steer someone else's weights toward a style, a character or a concept. They cannot run alone.
Family and weights: The base a model is built on, such as SD 1.5, SDXL, Flux or Wan. Every model inherits the release date of its family's weights.
Age: Degrees taken off for how long ago the family's weights came out: 150° a year, up to 750°. It keeps old downloads that had years to pile up from freezing the top of the list.
Family boost: A small, bounded nudge for LoRAs of a family this site is pushing forward right now, always listed in the breakdown.
Stand-in score: A stable number below the floor given to a model no source reports on yet, so lists stay complete. It never ranks above a model with data.
New! (grace period): A model less than 7 days old shows New! instead of a score and holds no place in the rankings. Its score is computed from day one, but a model one day old grows faster than anything, so without the wait the week's #1 would always be whatever came out yesterday rather than what held up for a week.
Trend: How a model's growth over a period compares with the typical model's growth over the same period. Faster than typical reads as a rise, slower as a fall.
Typical model: The middle of the catalog: half the models with a figure grew faster over the period, half slower.

How this model's popularity has moved against the rest of the catalog: its growth in downloads, likes and usage over each period, measured against the typical model's growth over the same period, with its heat score at the time. Above the typical growth reads as a rise, below it as a fall. The past is reconstructed from today's totals, the model's first day and its measured download speed where two readings exist; a model that did not exist at the start of a period shows n/a.

Similar Models

Qwen3 235B A22B Thinking 2507

TEXT / LLM

GPT 4o (2024 05 13)

TEXT / LLM

AI Models for Image, Video & Text Generation

Kimi K3 is a text model for the Text / LLM family listed on ModelTrends.ai, a read-only catalog of open source AI image, video and text models. Compare it with other Text / LLM models and check its heat score to see how it is trending.

More Text / LLM models llm moonshotai vision video reasoning coding agents long context

Browse by Model Family