Benchmark Almanac

K2-Think

K2-Think.

6
Benchmarks with a score
0
Ranked first, out of five or more models
0
In the top three
Mohamed bin Zayed University of Artificial Intelligence Sep 2025

K2-Think

Open weights Mid-size

No Epoch Capabilities Index score yet
  • BRUMO 2025 83.3% #33 /42
  • SMT 2025 79.7% #29 /40
  • CMIMC 2025 68.1% #29 /33

Where it ranks

Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.

Math and science

BenchmarkScoreRankSource
AIME 2025 83.3% #32 of 53 Official leaderboard
HMMT 2025 65% #34 of 53 Official leaderboard
SMT 2025 79.7% #29 of 40 Official leaderboard
BRUMO 2025 83.3% #33 of 42 Official leaderboard
CMIMC 2025 68.1% #29 of 33 Official leaderboard

Agents and tools

BenchmarkScoreRankSource
LLM Chess -458 #110 of 124 Official leaderboard