Skip to content

Kimi K2

Moonshot AIProprietaryNon-reasoningProvisionalReleased Jul 11, 2025Input: text
Compare with Mistral Large 3Source scores ↗OpenRouter ↗
Overall score
33.5
Rank 169 of 222 · 23.7–43.2
Context window
128K
128,000 tokens
Price per 1M tokens
$0.60 in · $2.50 out
Source: BenchLM
Blended price
$1.07
3:1 input:output tokens
Output speed
—
Not measured yet

Category scores

Scores per capability area with the model's rank among ranked models in that category.

  1. 1Reasoning61.4
  2. 2Instruction following#9648.4
  3. 3Coding40.8
  4. 4Knowledge#13140.2
Higher is better. #N is the model's rank in that category, where it is eligible to be ranked.

Benchmark results (14)

As reported in system cards, papers and public evaluations. Position compares this model with every other model reporting the same benchmark.

Knowledge

GPQA75.1#71 of 117
HLE4.7#69 of 69
MMLU-Pro81.1#34 of 67
MMLU89.5#7 of 28
SuperGPQA57.2#18 of 20
SimpleQA31.0#4 of 10

Math

AIME 202549.5#30 of 38
MATH-50097.4#4 of 20
AIME 202469.6#12 of 19

Coding

SWE-bench Verified65.8#70 of 98
LiveCodeBench v653.7#39 of 46

Multilingual

SWE Multilingual47.3#43 of 46

Multimodal

Design Arena Website1050#82 of 91

Instruction following

IFEval89.8#22 of 53

Pricing by source

USD per 1M tokens.

SourceInputOutputCached in
List price (BenchLM)$0.60$2.50—
OpenRouter moonshotai/kimi-k2$0.57$2.30—

Last published rate. Moonshot's Kimi platform pricing page listed Kimi K2 at $0.60 input / $2.50 output per million tokens, with cache-hit input at $0.15. The chat pricing page, read September 25, 2026, no longer lists it.

Speed

Measured by us through OpenRouter (median of trials).

Not measured yet.

Related models

Similar score