Kimi K2

No ratings yet

Be the first to review this model

Moonshot AI's trillion-parameter MoE model that has quickly become one of the most capable open-weight models available. Strong performance across coding, reasoning, and multilingual tasks with competitive benchmark scores. A rising star in the open-source AI community with broad hosting availability.

Released

July 1, 2025

Parameters

1T (MoE)

Context

128K

Pricing

Open Source

Benchmark Performance

Benchmark	Category	Score	Performance
MMLU	knowledge	89.5%	90
HumanEval	coding	77.9%	78
MATH	reasoning	97.4%	97

Last updated: March 15, 2026

Benchmark scores may vary based on evaluation methodology and conditions.

Reviews(0)

No reviews yet

Be the first to share your experience with Kimi K2.