> Kimi K3 is an open-weights AI model from Moonshot, released 16 Jul 2026. Index score 172 (GPT-4 = 100), rank 13 of 229 ranked models. API price $0.68 in / $10 out per million tokens, 1.05M context.

[Models](https://themodelindex.org/models/) /[Moonshot](https://themodelindex.org/labs/moonshot/)

# Kimi K3

Moonshot

Released 16 Jul 2026

Open

Date verified

[Compare](https://themodelindex.org/compare/kimi-k3-vs-claude-opus-5/) [Run it yourself →](https://themodelindex.org/local/) [Announcement](https://www.kimi.com/blog/kimi-k3?ref=apifox.com)

172161–183

Index score

#13 of 229 models

4/ 8

Core tests taken

4 of 6 areas · plus 17 anchor tests

−13

Versus the frontier at release

best then: Claude Fable 5

$0.68/ $10

Price per million tokens

input / output, list price

1.05M

Context window

tokens it reads at once

## Where it sits · on the frontier

_Chart: Model Index score of every scored model since late 2022, with the best closed and best open model over time_

GPT-5.5 reached this score 3 months earlier.

## Test results · by area · tick = best by any model

Show 14 anchor tests

Reasoning & knowledge

evidence

GPQA Diamond

93%

Humanity's Last Exam

no result

Maths

evidence

FrontierMath T1-3

72%

Coding

evidence

SWE-bench Verified

no result

Terminal-Bench

no result

Agents

evidence

APEX-Agents

51%

Novel problems

evidence

ARC-AGI-2

60%

Long tasks

no evidence yet

METR time horizon

no result

Independent results from the Epoch AI benchmarking hub, best reasoning setting. The core tests are shown here; anchor tests also inform the score and can be shown above.

## What others say · rank on each leaderboard

The Model Index

independent tests, our method

#13

of 219 · 172

LMArena · Coding

coding questions in chat

#7

of 383 · 1541

LMArena · Maths

maths questions in chat

#15

of 373 · 1500

LMArena · Text

overall chat quality, judged by people

#16

of 388 · 1488

Epoch Capabilities Index

overall capability from many benchmarks

#13

of 239 · 157.6

LMArena · Creative writing

creative writing

#24

of 386 · 1460

Artificial Analysis Intelligence Index

overall capability across ten evaluations

#14

of 207 · 44

LMArena · WebDev

building web apps and front ends

#11

of 123 · 1658

Further left is better. Leaderboards measure different things: LMArena is people voting blind between two answers, the others are test scores. Click any row for the source.

## Best at · top three among current models

2nd

coding questions in chat

LMArena · Coding

Where this model places in the top three of today's models, on category leaderboards and on our core tests.

## Where to use it · prices checked 1 Oct 2026

#### From the lab

Kimi

app · subscription

↗

Moonshot API

API

↗

#### API providers · 19 · per million tokens, in / out

R

Relace

$0.69 / $10 · fp4

cheapest

M

Morph

$2.35 / $6.95 · fp8

↗

I

InferenceNet

$1.19 / $13 · fp4

↗

W

Wafer

$2.8 / $14

↗

M

Makora

$2.04 / $12.8

↗

P

Phala

$2.55 / $12.8

↗

D

DigitalOcean

$2.55 / $12.9

↗

S

Sail Research

$2.8 / $14 · fp4

↗

#### Run it yourself

🤗

Official weights

moonshotai/Kimi-K3

↗

Provider prices come from OpenRouter's live listing; going direct to a provider can differ. Subscriptions are the lab's own apps.

## Facts and sources

Released

16 Jul 2026

Announcement

[kimi.com ↗](https://www.kimi.com/blog/kimi-k3?ref=apifox.com)

Release entry

Kimi K3 (weights 07-27)

Scored as

Kimi K3

Artificial Analysis index

44 · max

Output speed

34 tokens/s

Hugging Face

[moonshotai/Kimi-K3 ↗](https://huggingface.co/moonshotai/Kimi-K3)

## Moonshot releases · around this one

Kimi K2.6

Open

20 APR 2026

index 156

Kimi K2.7 Code

Open

12 JUN 2026

index 153

Kimi K3

Open

16 JUL 2026

index 172

Kimi K2.8 Preview

Closed

11 SEP 2026

---
Source: https://themodelindex.org/models/kimi-k3/ · The Model Index · data as of 2026-10-01