MoonshotAI: Kimi K2 Thinking

MOONSHOTAI Developer Architecture Profile

OpenRouter ↗HuggingFace ↗

Intelligence (ELO)1426Chatbot Arena Verified

Max Context Limit262,144Max Out: 98,304 tokens

Prompt Caching75% OFF$0.15 / 1M cached

Standard API / 1M$3.10🧠 Deep Reasoner

Model Capabilities & Contract SLAs

Reasoning
Coding & Logic
Fictional
🧠 Extended Test-Time Reasoning

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...

Granular Pricing Matrix

Input Tokens (Standard Prompt)$0.60 / 1M

⚡ Cached Input (Prompt Cache Read)75% OFF$0.15 / 1M

Output Tokens (Completion)$2.50 / 1M

Pricing data via OpenRouter. Sync: 10/2/2026

Evaluate Competitors

VS Engine MatchupMoonshotAI: Kimi K2 Thinking vs Cohere: Command A+VS Engine MatchupMoonshotAI: Kimi K2 Thinking vs Z.ai: GLM 5.3 (batch)VS Engine MatchupMoonshotAI: Kimi K2 Thinking vs Amazon: Nova 2 Lite VS Engine MatchupMoonshotAI: Kimi K2 Thinking vs Z.ai: GLM 4.6 VS Engine MatchupMoonshotAI: Kimi K2 Thinking vs OpenAI: GPT-5 VS Engine MatchupMoonshotAI: Kimi K2 Thinking vs Z.ai: GLM Flash Latest