Claude 3.7 Sonnet (Reasoning)
AnthropicClaude
描述
The most intelligent Claude model and the first hybrid reasoning model on the market. Claude 3.7 Sonnet can produce near-instant responses or extended, step-by-step thinking that is made visible to the user. Shows particularly strong improvements in coding and front-end web development.
发布日期
2025-02-24
参数规模
—
上下文长度
200K
支持模态
image, pdf, text
能力雷达图
42
general
41
coding
62
reasoning
51
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
35.2%自报
Biology
GPQA
84.8%自报
Code
SWE-Bench Verified
70.3%自报
Communication
TAU-bench Retail
81.2%自报
TAU-bench Airline
58.4%自报
General
IFEval
93.2%自报
MMMLU
86.1%自报
MMMU
75.0%自报
Math
MATH-500
96.2%自报
AIME 2024
80.0%自报
AIME 2025
54.8%自报
AA 评测指数
Math Index56.3
Coding Index36.4
Intelligence Index27.1
Math 5000.9
Mmlu Pro0.8
Gpqa0.8
Lcr0.6
Aime 250.6
Tau20.5
Aime0.5
Ifbench0.5
Livecodebench0.5
Scicode0.4
Terminalbench Hard0.2
Hle0.1
LLM Stats 分类评分
Instruction Following90
Language90
Structured Output90
Math80
Multimodal80
Physics80
General80
Healthcare80
Biology80
Chemistry80
Vision80
Reasoning70
Frontend Development70
Communication70
Tool Calling70
Code50
Agents40
定价
输入价格免费
输出价格免费
混合价格(3:1)免费
缓存读取价格$0.3 / 1M tokens
缓存写入价格$3.75 / 1M tokens
速度
Tokens/秒0.0
首Token延迟0.00s
首回答延迟0.00s
供应商价格排行
供应商价格排行
3 个供应商
最便宜: Abacus最贵: Anthropic
供应商输入输出
1Abacus最便宜
$3
$15
2LLM Gateway
$3
$15
3Anthropic
$3
$15
比较该模型在不同 API 供应商之间的定价。