Home/Models/Kimi K2 Thinking
moonshotai logo

Kimi K2 Thinking

by moonshotai · Released Nov 2025

Open Source
61.0
avg score
Rank #64
Compare
Better than 73% of all models
Context
262K tokens (~131 books)
Input $/1M
$0.60
Output $/1M
$2.50
Type
text
License
Open Source
Benchmarks
25 tested
Data updated today
About

Kimi K2 Thinking is Moonshot AI’s most advanced open reasoning model to date, extending the K2 series into agentic, long-horizon reasoning. Built on the trillion-parameter Mixture-of-Experts (MoE) architecture introduced in...

Tested on 25 benchmarks with 53.3% average. Top scores: OpenCompass — AIME2025 (94.1%), OpenCompass — IFEval (92.4%), OpenCompass — MMLU-Pro (84.3%).

Looking for similar performance at lower cost?
gpt-oss-20b (free) scores 61.0 (100% as good) at $0.00/1M input · 100% cheaper
Capabilities
coding
54.1
#55 globally
reasoning
57.9
#35 globally
math
55.9
#60 globally
knowledge
48.5
#108 globally
agentic
4.0
#34 globally
language
73.6
#63 globally
Benchmark Scores
Compare All
Tested on 25 benchmarks · Ranked across 6 categories
Score Distribution (all 233 models)
0255075100
▲ You are here
OpenCompass — LiveCodeBenchV6

OpenCompass Live Code Bench v6. Fresh competitive programming problems to evaluate code generation without memorization.

77.1
LiveBench — Coding

Regularly refreshed coding problems that avoid data contamination. New problems added monthly to prevent memorization.

67.4
SWE-Bench Verified (Bash Only)

SWE-bench Verified solved using only bash commands, no specialized frameworks. Tests raw terminal-based problem solving.

63.4
LiveBench — Reasoning

Regularly refreshed reasoning problems testing logical deduction, spatial reasoning, and analytical thinking.

63.5
LiveBench — Data Analysis

Fresh data analysis tasks testing ability to interpret tables, charts, and statistical data.

52.3
OpenCompass — AIME2025

OpenCompass evaluation on AIME 2025 problems. Tests mathematical reasoning on fresh competition problems.

94.1
OTIS Mock AIME 2024-2025

Mock AIME (American Invitational Mathematics Exam) problems from OTIS. Tests mathematical competition performance.

83.0
LiveBench — Mathematics

Regularly updated math problems that test numerical reasoning, algebra, calculus, and combinatorics.

81.1
Excellent (85+) Good (70-85) Average (50-70) Below (<50)
Links
Documentation
Community
BenchGecko API
kimi-k2-thinking
Specifications
  • Typetext
  • Context262K tokens (~131 books)
  • ReleasedNov 2025
  • LicenseOpen Source
  • StatusActive
  • Cost / Message~$0.004
Available On
moonshotai logomoonshotai$0.60
Share & Export
Tweet
Kimi K2 Thinking is an open-source text AI model by moonshotai, released in November 2025. It has an average benchmark score of 61.0. Context window: 262K tokens.