Claude Opus 4.6 (Adaptive Reasoning, Max Effort)
AnthropicClaude
描述
Claude Opus 4 is Anthropic's most powerful model and the world's best coding model, part of the Claude 4 family. It delivers sustained performance on complex, long-running tasks and agent workflows. Opus 4 excels at coding, advanced reasoning, and can use tools (like web search) during extended thinking. It supports parallel tool execution and has improved memory capabilities.
发布日期
2026-02-05
参数规模
—
上下文长度
1.0M
支持模态
image, pdf, text
能力雷达图
42
general
52
coding
90
reasoning
66
science估算
70
agents
80
multimodal
Science 在缺少专门科学评测时使用推理能力代理估算。
排行榜排名
基准测试分数 (LLM Stats)
Agents
Terminal-Bench
39.2%自报
Biology
GPQA
79.6%自报
Code
SWE-Bench Verified
72.5%自报
Communication
TAU-bench Retail
81.4%自报
TAU-bench Airline
59.6%自报
General
MMMLU
88.8%自报
MMMU (validation)
76.5%自报
Math
AIME 2025
75.5%自报
Reasoning
ARC-AGI v2
8.6%自报
AA 评测指数
Intelligence Index43.7
Tau20.9
Gpqa0.9
Lcr0.7
Ifbench0.5
Scicode0.5
Terminalbench Hard0.5
Hle0.4
LLM Stats 分类评分
Language90
Math80
Multimodal80
Physics80
General80
Healthcare80
Biology80
Chemistry80
Frontend Development70
Communication70
Tool Calling70
Reasoning60
Code60
Agents40
Vision40
Spatial Reasoning10
定价
输入价格$5 / 1M tokens
输出价格$25 / 1M tokens
混合价格(3:1)$10 / 1M tokens
缓存读取价格$0.5 / 1M tokens
缓存写入价格$6.25 / 1M tokens
速度
Tokens/秒66.0
首Token延迟17.46s
首回答延迟17.46s
供应商价格排行
供应商价格排行
1 个供应商
供应商输入输出
1Anthropic主要
$5
$25
比较该模型在不同 API 供应商之间的定价。