Benchmark Almanac

Moonshot AI

Best on the Capabilities Index Kimi K3.1 Pro 161.4
15 models 14 model cards read moonshot.ai

How does Moonshot AI stack up?

Every model on Epoch AI's Capabilities Index by release date, with Moonshot AI's in color.

50 100 150 2023 2024 2025 2026 2027 Kimi K3.1 Pro Kimi K2.7 Code Kimi K2 Thinking 50 100 150 2023 2024 2025 2026 2027 Kimi K3.1 Pro Kimi K2.7 Code

Source: Epoch Capabilities Index, Epoch AI (CC BY 4.0)

Moonshot AI's best models

Moonshot AI Oct 2026

Kimi K3.1 Pro

Open weights

161.4 #16 of 256
Epoch Capabilities Index
Self-Reported Index 170.0 #4 of 404
  • MMMU-Pro 84.1% #2 /90
  • Terminal-Bench 4.0 60.9% #3 /40
  • GPQA Diamond 94.2% #12 /229
Moonshot AI Jul 2026

Kimi K3

Open weights Flagship

158.6 #24 of 256
Epoch Capabilities Index
Self-Reported Index 159.5 #35 of 404
  • MMMU-Pro 81.6% #10 /90
  • DeepSWE 68.5% #8 /28
  • AutomationBench 22.7% #14 /42
Moonshot AI Apr 2026

Kimi K2.6

Open weights Flagship

151.6 #62 of 256
Epoch Capabilities Index
Self-Reported Index 153.6 #69 of 404
  • SWE-bench Verified 76.7% #10 /32
  • MMMU-Pro 79.4% #19 /90
  • AutomationBench 2.59% #38 /42
Moonshot AI Jun 2026

Kimi K2.7 Code

Open weights Mid-size

151.2 #64 of 256
Epoch Capabilities Index
  • DeepSWE 30.5% #26 /28
  • AutomationBench 3.20% #37 /42
  • GPQA Diamond 87.9% #66 /229
Moonshot AI Jan 2026

Kimi K2.5

Open weights Flagship

148.8 #75 of 256
Epoch Capabilities Index
Self-Reported Index 149.6 #99 of 404
  • Humanity's Last Exam 24.4% #17 /47
  • SWE-bench Verified 73.8% #18 /32
  • MMMU-Pro 78.5% #23 /90
Moonshot AI Nov 2025

Kimi K2 Thinking

Open weights Flagship

145.4 #93 of 256
Epoch Capabilities Index
Self-Reported Index 146.8 #117 of 404
  • Terminal-Bench 2.0 35.7% #27 /43
  • SciCode 44.8% #19 /49
  • MultiChallenge 55.4% #25 /42

What benchmarks does Moonshot AI mention most?

Here are the 28 benchmarks Moonshot AI mentions most in its launch posts.

Source: Labs’ launch posts, read by Benchmark Almanac

Model cards

DateModelDocumentBenchmarks
Oct 24, 2026 Kimi K3.1 Pro Kimi K3.1 Pro Tech Blog: Advancing Open Intelligence 38
Jul 16, 2026 Kimi K3 Kimi K3 model card (Hugging Face) 45
Jun 12, 2026 Kimi K2.7 Code Kimi K2.7 Code model card (Hugging Face) 6
Apr 20, 2026 Kimi K2.6 Kimi K2.6 model card (Hugging Face) 25
Jan 27, 2026 Kimi K2.5 Kimi K2.5 model card (Hugging Face) 38
Nov 6, 2025 Kimi K2 Thinking Kimi K2 Thinking model card (Hugging Face) 21
Sep 5, 2025 Kimi K2 Sep 2025 Kimi K2-Instruct-0905 model card (Hugging Face) 5
Jul 11, 2025 Kimi K2 Jul 2025 Kimi K2 Instruct model card (Hugging Face) 27
Jun 21, 2025 Kimi-VL-A3B-Thinking-2506 Kimi-VL-A3B-Thinking-2506 model card (Hugging Face) 21
Jun 16, 2025 Kimi-Dev-72B Introducing Kimi-Dev: A Strong and Open-source Coding LLM for Issue Resolution (model card) 1
Apr 25, 2025 Kimi-Audio-7B-Instruct Kimi-Audio GitHub README (evaluation section) and technical report 9
Apr 10, 2025 Kimi-VL-A3B-Thinking Kimi-VL-A3B-Thinking model card (Hugging Face) and Kimi-VL Technical Report 3
Feb 22, 2025 Moonlight-16B-A3B Moonlight-16B-A3B model card (Hugging Face), Muon is Scalable for LLM Training 11
Jan 20, 2025 Kimi k1.5 Kimi k1.5: Scaling Reinforcement Learning with LLMs (tech report, arXiv 2501.12599, Tables 2 and 3) 12

Every Moonshot AI model