ModelBeat
Models/Kimi K2
All models
M
moonshotai/kimi-k2

Kimi K2 is Moonshot’s open-weights AI model, released Sep 5, 2025. It costs $0.57/M input and $2.30/M output tokens, and handles a 131K-token context window.

Kimi K2 Instruct is a large-scale Mixture-of-Experts (MoE) language model developed by Moonshot AI, featuring 1 trillion total parameters with 32 billion active per forward pass.

Context
131K
Input
$0.57 / 1M
Output
$2.30 / 1M
Released
Sep 5, 2025
Modalities
texttext

Benchmarks

Scores on standardized evaluations. Higher is better — and rank shows where Kimi K2 lands among all models tracked on Model Beat.

13.0
Intelligence Index
Epoch AI
13th percentile of tracked models
9.0
Coding Index
Epoch AI
9th percentile of tracked models
21.0
Agentic Index
Epoch AI
21st percentile of tracked models

Reasoning

5 evals
GPQA Diamond76.6%

Graduate-level scientific reasoning

Humanity's Last Exam7.0%

Frontier of human expert knowledge

MMLU-Pro82.4%

Broad expert knowledge

SimpleBench26.3%

Common-sense trick questions

WeirdML39.4%

Novel ML problem-solving

Coding

3 evals
GSO (code optimization)4.9%

Code performance optimization

LiveCodeBench55.6%

Contamination-free coding

SciCode34.5%

Scientific research coding

Math

1 evals
AIME 2024/202557.0%

Olympiad-qualifier math

Agentic & Tools

2 evals
Terminal-Bench27.8%

Command-line agentic tasks

τ²-bench61.1%

Tool-agent-user reliability

In the news

17

Compare Kimi K2 with

Frequently asked questions

What is Kimi K2?

Kimi K2 Instruct is a large-scale Mixture-of-Experts (MoE) language model developed by Moonshot AI, featuring 1 trillion total parameters with 32 billion active per forward pass.

Who created Kimi K2?

Kimi K2 is an AI model developed by Moonshot, part of the Kimi family, released Sep 5, 2025.

How much does Kimi K2 cost?

Kimi K2 is served from $0.57 per million input tokens and $2.30 per million output tokens (cheapest credible provider on OpenRouter: Novita).

What is Kimi K2's context window?

Kimi K2 supports a context window of up to 131K tokens.

How does Kimi K2 perform on benchmarks?

On standardized evaluations tracked by Epoch AI, Kimi K2 scores 82.4% on MMLU-Pro, 61.1% on τ²-bench, 76.6% on GPQA Diamond.

Is Kimi K2 open source?

Kimi K2 is released with open weights (Open weights (restricted use)), so the model can be downloaded and self-hosted.

Benchmark data from Epoch AI (CC BY) and Artificial Analysis; pricing, specs & descriptions from OpenRouter. Data updated Jul 30, 2026.