MoonshotAI: Kimi K2 Thinking
moonshotai/kimi-k2-thinking
78th smartest of 182ReasoningToolsJSON
Intelligence
32.7
78th of 182
Design Elo
1137
Website
Speed
—
tokens/sec
Latency
—
first token
Input price
$0.600
184th cheapest
Context
262K
100K max out
How it compares
Smarter than57%
of all ranked models
Cheaper than45%
of all ranked models
Overview
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...
Benchmarks
independent · Artificial Analysis & Design ArenaArtificial Analysis79th percentile
Intelligence Index
32.7
GPQA Diamond
84%
Humanity's Last Exam
22%
SciCode
42%
Tau²-Bench (agentic)
93%
Design Arena · Elo713 tournaments
Website
1137
Providers & pricing (2)
| Provider | In $/M | Out $/M | Context | Uptime |
|---|---|---|---|---|
| Novitabf16 | $0.600 | $2.50 | 262K | 99.7% |
| $0.600 | $2.50 | 262K | 100% |
Specifications
Context window262K
Max output100K
Knowledge cutoff—
Input modalitiestext
Output modalitiestext
Prompt caching—
Cache read price$0.150/M
ModeratedNo
Open weightsmoonshotai/Kimi-K2-Thinking ↗
Kimi K2 Thinking FAQ
How much does Kimi K2 Thinking cost?
Kimi K2 Thinking costs $0.600 per million input tokens and $2.50 per million output tokens via OpenRouter, making it 184th cheapest of 332 paid models.
How smart is Kimi K2 Thinking?
Kimi K2 Thinking scores 32.7 on the Artificial Analysis Intelligence Index, ranking 78th of 182 benchmarked models, with a GPQA Diamond score of 84%.
What is Kimi K2 Thinking's context window?
Kimi K2 Thinking supports a 262K-token context window and can output up to 100K tokens. It accepts text input.
Similar models
All Kimi K2 Thinking alternatives →Compare head-to-head
MoonshotAI: Kimi K2 Thinking vs OpenAI: o3 ProMoonshotAI: Kimi K2 Thinking vs Qwen: Qwen3.5-122B-A10BMoonshotAI: Kimi K2 Thinking vs Meituan: LongCat 2.0MoonshotAI: Kimi K2 Thinking vs Tencent: Hy3 previewMoonshotAI: Kimi K2 Thinking vs Kwaipilot: KAT-Coder-Pro V2MoonshotAI: Kimi K2 Thinking vs Qwen: Qwen3.5 397B A17B