in the loop

Kimi K2 Thinking
#79 of 274 scored.

MoonshotNov 6, 2025Open weights

Capabilities Index

Range 143.4–147.5

146.0

Context

Takes text

262K

Input

Per 1M tokens

$0.60

Output

Per 1M tokens

$2.50

Benchmarks

How it scores.

  • GPQA Diamond79.0%

    67th of 186

  • Mock AIME83.0%

    76th of 176

  • Terminal-Bench35.7%

    22nd of 35

Everything else Epoch has for it

  • METR Time Horizons59.2%
  • WeirdML42.8%
  • FrontierMath-2025-02-28-Private37.5%
  • CL-bench17.6%
  • Chess Puzzles15.8%
  • FrontierMath-Tier-4-2025-07-01-Private0.0%

On the index

Around it.

  1. #76DeepSeek-V3.2DeepSeek · $0.26 in146.3
  2. #77Nemotron 3 UltraNvidia · $0.50 in146.2
  3. #78DeepSeek-V4-FlashDeepSeek · $0.006 in146.1
  4. #79Kimi K2 ThinkingMoonshot · $0.60 in146.0
  5. #80MiniMax-M2.7MiniMax · $0.21 in145.8
  6. #81GLM-5Z.ai · $0.60 in145.8
  7. #82GPT-5.4 NanoOpenAI · $0.20 in145.8

Sources