Claude 3.7 Sonnet Thinking
LatestAnthropic
•
Proprietary# 75
Released
Feb 24, 2025
# 9
Knowledge Cutoff
Oct 24
# 9
Context Length
200K
Benchmarks
# 93
Code RankedAGI
48.6%
# 99
Agentic RankedAGI
38.4%
# 10
Coding LiveBench 25.5
73.2%
# 14
Code LMArena
1333
# 13
Aider Polyglot
64.9%
# 5
Code LiveBench (old)
71.5%
# 104
Reason RankedAGI
43.0%
# 57
HLE
8.9%
# 52
GPQA Diamond
78.2%
# 13
Reason LiveBench 25.5
76.2%
# 44
Text Arena
1363
# 54
AIME 2025 I & II
49.5%
# 18
AIME 2024
80.0%
# 12
Math LiveBench 25.5
79.0%
# 17
NYT Connections
33.6%
# 15
MMMU
75.0%
# 7
IF LiveBench 25.5
81.3%
# 11
Avg LiveBench 25.5
66.9%
# 3
Avg LiveBench (old)
74.3%
# 2
IF Evaluation
93.2%
# 24
Coding LiveBench 25.4
44.7%
# 2
Data LiveBench
72.8%
# 6
Language LiveBench
61.0%
# 6
Agentic LiveBench 25.5
25.0%
# 109
Math RankedAGI
55.7%
# 100
RAGI RankedAGI
46.4%
# 36
GDPval AA
1054
Pricing
# 38
Input Cost /M
$3
# 44
Output Cost /M
$15