Codesota · Models · Kimi K2-Thinking-0905Moonshot AI9 results · 3 benchmarks
Model card

Kimi K2-Thinking-0905.

Moonshot AIopen-source
§ 01 · Benchmarks

Every benchmark Kimi K2-Thinking-0905 has a recorded score for.

#BenchmarkArea · TaskMetricValueRankDateSource
01MMLU-ProReasoning · Commonsense Reasoningaccuracy84.6%#18/202026-04-20source ↗
02SWE-bench VerifiedAgentic AI · SWE-benchresolve-rate71.3%#36/81source ↗
03PLCCNatural Language Processing · Polish Cultural Competencygrammar73.0%#39/165source ↗
04PLCCNatural Language Processing · Polish Cultural Competencygeography84.0%#45/165source ↗
05PLCCNatural Language Processing · Polish Cultural Competencyaverage71.7%#54/165source ↗
06PLCCNatural Language Processing · Polish Cultural Competencyart-and-entertainment63.0%#54/165source ↗
07PLCCNatural Language Processing · Polish Cultural Competencyhistory80.0%#62/165source ↗
08PLCCNatural Language Processing · Polish Cultural Competencyculture-and-tradition71.0%#66/165source ↗
09PLCCNatural Language Processing · Polish Cultural Competencyvocabulary59.0%#75/165source ↗
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 02 · Strengths by area

Where Kimi K2-Thinking-0905 actually performs.

Reasoning
1
benchmark
avg rank #18.0
Agentic AI
1
benchmark
avg rank #36.0
Natural Language Processing
1
benchmark
avg rank #56.4
§ 04 · Related models

Other Moonshot AI models scored on Codesota.

Kimi K2.5
Undisclosed params · 5 results
Kimi K2-Instruct-0905
1 result
Kimi K2.5
1 result
Kimi k1.5
1 result
Kimi-Audio
0 results
§ 05 · Sources & freshness

Where these numbers come from.

sdadas/PLCC
7
results
llm-stats
1
result
editorial
1
result
8 of 9 rows marked verified.