Claude Instant
AnthropicClaude
发布日期
2023-03-14
参数规模
—
上下文长度
1.0M
支持模态
image, pdf, text
能力雷达图
15
general
9
coding
15
reasoning
24
science估算
80
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
Vending-Bench 2
801759.0%自报
GDPval-AA
1606.00 / 3000自报
DeepSearchQA
91.3%自报
BrowseComp
84.0%自报
CyberGym
73.8%自报
OSWorld
72.7%自报
Terminal-Bench 2.0
65.4%自报
MCP Atlas
62.7%自报
Finance Agent
60.7%自报
FrontierSWE
56.0%自报
OpenRCA
34.9%自报
Legal Agent Benchmark
4.2%自报
Biology
GPQA
91.3%自报
Code
SWE-Bench Verified
80.8%自报
SWE-bench Multilingual
77.8%自报
Communication
Tau2 Telecom
99.3%自报
Tau2 Retail
91.9%自报
General
MMMLU
91.1%自报
MMMU-Pro
77.3%自报
LiveBench
76.3%自报
MRCR v2 (8-needle)
76.0%自报
Healthcare
FigQA
78.3%自报
Long Context
Graphwalks parents >128k
95.4%自报
Graphwalks BFS >128k
61.5%自报
Math
AIME 2025
99.8%自报
Humanity's Last Exam
53.1%自报
Multimodal
CharXiv-R
77.4%自报
Reasoning
ARC-AGI v2
68.8%自报
AA 评测指数
Coding Index7.8
Intelligence Index2.1
Mmlu Pro0.4
Gpqa0.3
Math 5000.3
Livecodebench0.1
Hle0.0
Aime0.0
LLM Stats 分类评分
Agents100
Finance100
Legal100
Reasoning100
General100
Communication100
Language90
Physics90
Search90
Biology90
Chemistry90
Long Context80
Math80
Multimodal80
Safety80
Spatial Reasoning80
Frontend Development80
Healthcare80
Tool Calling80
Code70
Vision70
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$0.5 / 1M tokens
缓存写入价格$6.25 / 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
暂无提供商数据