趋近智
参数
-
上下文长度
200K
模态
Text
架构
Dense
许可证
Proprietary
发布日期
1 Nov 2025
训练数据截止日期
-
注意力结构
Multi-Head Attention
隐藏维度大小
-
层数
-
注意力头
-
键值头
-
激活函数
-
归一化
-
位置嵌入
Absolute Position Embedding
Anthropic's most powerful Claude model with maximum thinking effort configuration. Exceptional performance across all domains with top-tier web development (Elo 1510), reasoning (80.09), and mathematics (90.39). Features extended reasoning chains and deep analytical capabilities. Ideal for complex problems requiring thorough analysis, multi-step reasoning, and comprehensive solutions. Extended context window up to 200K tokens.
Enhanced Claude models with further improvements in reasoning, coding, and agentic capabilities. Features advanced thinking modes with adjustable effort levels (high, medium, standard) for optimal performance-latency tradeoffs. Excels at complex analysis, software development, web development, and long-context understanding. Includes thinking variants that expose reasoning process for improved transparency.
排名
#2
| 基准 | 分数 | 排名 |
|---|---|---|
Coding Aider Coding | 0.84 | 🥇 1 |
Web Development WebDev Arena | 1510 | 🥇 1 |
Agentic Coding LiveBench Agentic | 0.70 | 🥈 2 |
Coding LiveBench Coding | 0.80 | ⭐ 5 |
Reasoning LiveBench Reasoning | 0.80 | ⭐ 7 |
StackUnseen ProLLM Stack Unseen | 0.82 | 8 |
Mathematics LiveBench Mathematics | 0.77 | 14 |
排名
#2 🥈
编程排名
#1 🥇