Kimi K2 Thinking
by Moonshot AIRank #222Score 53.3
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...
性能概览
评分
53.3
排名
#222
24小时变化
0
7天变化
-64
状态
stable置信度
high信号评分
| 信号 | 标准化 | 权重 | 贡献 | 新鲜度 |
|---|---|---|---|---|
Capabilities capability | 66.7 | 20% | 13.3 | 2026-08-08T13:39:50.253Z |
Benchmarks benchmark | 50.7 | 30% | 15.2 | 2026-08-08T13:39:50.253Z |
Pricing pricing_tier | 97.5 | 15% | 14.6 | 2026-08-08T13:39:50.253Z |
Context Window context_window | 86.0 | 10% | 8.6 | 2026-08-08T13:39:50.253Z |
Recency recency | 82.7 | 15% | 12.4 | 2026-08-08T13:39:50.253Z |
Output Capacity output_capacity | 83.1 | 10% | 8.3 | 2026-08-08T13:39:50.253Z |
主要驱动因素
neutral
Benchmarks
positive
Pricing
$2.50/M output tokens
$2.50
positive
Capabilities
Supports reasoning, tools, JSON mode, streaming
4/7
positive
Recency
Released over 1 year ago
9mo ago
功能
| 功能 | 支持 |
|---|---|
| 视觉 | 否 |
| 推理 | 是 |
| JSON模式 | 是 |
| 流式输出 | 是 |
| 函数调用 | 是 |
| 网页搜索 | 否 |
价格
输入/百万令牌
$0.60
输出/百万令牌
$2.50
上下文窗口
262K
最大输出
100K