Kimi K2.6
Where it stands
Quality in independent tests against price, next to 135 other models. Up means more capable, left means cheaper.
About this model
Kimi K2.6 is an AI model from Moonshot AI, made available through the API on April 20, 2026. It can read up to 262,144 tokens in one request, roughly 492 pages of text, and write up to 235,929 tokens in a single answer. Besides text, it accepts images as input.
Through the API it costs $0.65 per million input tokens and $3.41 per million output tokens. Its pricing sits in the middle of the market.
It supports extended reasoning: it can "think" step by step before answering, which helps on maths, code and complex questions. It can call tools and functions, so developers can connect it to search, databases or other software. Its weights are openly published (moonshotai/Kimi-K2.6), so it can also be downloaded and run on your own hardware.
Quality scores
Rankings →- SWE-bench VerifiedCan the model fix real bugs in real open-source software projects? Share of tasks solved.76.7%#9 of 28
- GPQA DiamondVery hard multiple-choice questions in biology, physics and chemistry, written by PhD experts. Random guessing scores 25%.90.8%#26 of 116
- AIME mathCompetition-level math problems in the style of the American Invitational Mathematics Examination. Share solved.96.1%#21 of 112
- FrontierMathOriginal, unpublished math problems written by professional mathematicians, from advanced to research level. Share solved.57.2%#35 of 73
- SimpleQA VerifiedShort factual questions. Measures how often the model gets facts right instead of making something up.34.9%#44 of 69
Where to use it
Companies that serve this model through an API: 18. Prices per million tokens, cheapest first.
| Provider | Inputi | Outputi | Contexti | Uptime (24 h)i |
|---|---|---|---|---|
InceptronCheapestint4 | $0.47 | $2.45 | 256K | 99.1% |
DigitalOcean | $0.57 | $2.40 | 256K | 98.5% |
Decartfp4 | $0.59 | $2.47 | 256K | 97.8% |
StreamLakefp8 | $0.60 | $2.52 | 256K | 99.1% |
Chutesint4 | $0.50 | $2.85 | 256K | 99.6% |
CoreWeavefp4 | $0.65 | $3.41 | 256K | 97.6% |
SiliconFlowfp8 | $0.77 | $3.40 | 256K | 99.8% |
Crusoebf16 | $0.70 | $3.50 | 256K | 89.8% |
Novita | $0.80 | $3.40 | 256K | 99.8% |
DeepInfrafp4 | $0.75 | $3.50 | 256K | 98.8% |
Veniceint4 | $0.75 | $3.50 | 256K | 94.2% |
Parasailint4 | $0.75 | $3.50 | 256K | 99.3% |
GMICloudfp8 | $0.86 | $3.60 | 256K | — |
Baidufp4 | $0.95 | $4 | 256K | 100.0% |
AtlasCloudint4 | $0.95 | $4 | 256K | 98.8% |
Cloudflare | $0.95 | $4 | 256K | 99.7% |
Moonshot AIOfficialint4 | $0.95 | $4 | 256K | 99.4% |
Phala | $1.09 | $4.60 | 256K | 99.8% |
Availability and prices come from OpenRouter’s public API and are refreshed daily. Uptime is the share of successful requests over the last 24 hours. Some providers run a compressed version of open models (fp8, int4), which is cheaper but can be slightly less accurate.
Use it in your code
Call Kimi K2.6 through OpenRouter’s OpenAI-compatible API: create a key on openrouter.ai, then run one of these examples.
curl https://openrouter.ai/api/v1/chat/completions \
-H "Authorization: Bearer $OPENROUTER_API_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "moonshotai/kimi-k2.6", "messages": [{"role": "user", "content": "Hello! What can you do?"}]}'import os
from openai import OpenAI
client = OpenAI(base_url="https://openrouter.ai/api/v1", api_key=os.environ["OPENROUTER_API_KEY"])
reply = client.chat.completions.create(
model="moonshotai/kimi-k2.6",
messages=[{"role": "user", "content": "Hello! What can you do?"}],
)
print(reply.choices[0].message.content)import OpenAI from "openai";
const client = new OpenAI({ baseURL: "https://openrouter.ai/api/v1", apiKey: process.env.OPENROUTER_API_KEY });
const reply = await client.chat.completions.create({
model: "moonshotai/kimi-k2.6",
messages: [{ role: "user", content: "Hello! What can you do?" }],
});
console.log(reply.choices[0].message.content);The same code works with every model on this site: only the model ID changes. Many labs also offer their own API and SDK.
What it costs in practice
- 1,000 chat replies$1.35
- Summarise a 100-page document$0.036
- An app with 10,000 requests a day for a month$604
Similar-priced alternatives
| Model | Inputi | Outputi | Contexti | |
|---|---|---|---|---|
DeepSeek V4 Pro 0813DeepSeek | $0.66 | $1.98 | 1M | Compare with ⇄ |
Aion 3.5 MiniAion Labs | $0.70 | $1.40 | 256K | Compare with ⇄ |
Nemotron 3 UltraNVIDIA | $0.60 | $2.40 | 256K | Compare with ⇄ |
KAT-Coder-Pro V2.5Kuaishou | $0.74 | $2.96 | 256K | Compare with ⇄ |
Gemini 3.8 FlashGoogle DeepMind | $0.75 | $3.75 | 1M | Compare with ⇄ |
Hy4 previewTencent | $0.83 | $2.50 | 1M | Compare with ⇄ |
News about Moonshot AI
See all →Frequently asked questions
- How much does Kimi K2.6 cost?
- Kimi K2.6 costs $0.65 per million input tokens and $3.41 per million output tokens through the API. For example, 1,000 chat replies of about 500 tokens in and 300 out cost around $1.35.
- What is the context window of Kimi K2.6?
- Kimi K2.6 can read up to 262,144 tokens in one request, about 492 pages of text. It can write up to 235,929 tokens in a single answer.
- Is Kimi K2.6 open source?
- Yes. Its weights are published (moonshotai/Kimi-K2.6), so you can download it and run it on your own hardware, or use it through one of 18 API providers.
- Who makes Kimi K2.6?
- Kimi K2.6 is made by Moonshot AI. It became available through the API on April 20, 2026.
- How good is Kimi K2.6?
- Kimi K2.6 has an overall index of 151.0 in Epoch AI’s independent tests, ranking #41 of 136 models we track.
- Can Kimi K2.6 read images?
- Yes. Besides text, it accepts images as input.
- What are the alternatives to Kimi K2.6?
- Models with a similar price from other companies include DeepSeek V4 Pro 0813, Aion 3.5 Mini and Nemotron 3 Ultra. You can compare them side by side on this page.