Moonshot AI · all models ›

MKimi K2 0905OPEN

Kimi K2 0905 is the September update of Kimi K2 0711.

262KContext window · tokens
$0.6Input price · per M tokens
$2.5Output price · per M tokens
Moonshot AIProvider

Prices update automatically — checked daily against provider list prices.

See model comparisons → Compare all model prices

Benchmarks & performance

Independent benchmark scores for Kimi K2 0905, measured by Artificial Analysis. Higher is better.

3.5×
More intelligence per dollar than Claude Opus 5 (Fast).
For the same budget, Kimi K2 0905 delivers 3.5 times more capability. Open weights also mean you can self-host it and pay nothing per token.
Intelligence index24
Math index57.3
GPQA76.7%
MMLU-Pro81.9%
Humanity's Last Exam6.4%
Long Context Reasoning53.7%
LiveCodeBench61%
SciCode30.7%
AIME 202557.3%
IFBench41.7%
τ²-Bench73.4%
Terminal-Bench Hard23.5%
💰 Blended price$1.075 / 1M tokens
📈 Value22.3 intelligence points per $
vs. models measured heretop 57%
Scores higher than 43% of the 273 models measured by Artificial Analysis and tracked here.
Models at this level cost $0.56 per 1M tokens (median of 15) — this one costs $1.08.
Cheaper and better on this index: GLM 5.3 Flash · Gemini 3.7 Flash · Gemini 3.7 Flash (batch) and 56 more
Benchmark data by Artificial Analysis

About this model

Kimi K2 0905 is an open-weight AI model by Moonshot AI. You can download and self-host it for free; the prices below are hosted-API list prices, tracked daily, for when you prefer convenience over self-hosting.

Frequently asked questions

What is Kimi K2 0905?

Kimi K2 0905 is an AI language model from Moonshot AI. It is open-weight: you can download it and run it on your own hardware, for free. It scores 24 on the Artificial Analysis intelligence index.

Is Kimi K2 0905 free?

The weights are free and open — you can self-host Kimi K2 0905 and pay nothing per token. If you prefer a hosted API, list prices are $0.6 per million input tokens and $2.5 per million output tokens.

What is Kimi K2 0905 good at?

Independent benchmarks from Artificial Analysis give it GPQA 76.7%, MMLU-Pro 81.9%, Humanity's Last Exam 6.4%, Long Context Reasoning 53.7%, LiveCodeBench 61%, SciCode 30.7%, AIME 2025 57.3%, IFBench 41.7%, τ²-Bench 73.4%, Terminal-Bench Hard 23.5%. It is particularly used for mathematical reasoning.

Can I self-host Kimi K2 0905?

Yes. Kimi K2 0905 has open weights, so you can download it and run it on your own GPU or server with tools like Ollama, vLLM or llama.cpp — with no per-token cost.

Related models

Kimi K3Moonshot AIKimi K2.6Moonshot AIKimi K2.5Moonshot AIKimi K2.7 CodeMoonshot AIKimi K2 0711Moonshot AIKimi K2 ThinkingMoonshot AI