Claude 3.5 Sonnet
LatestAnthropic
•
Proprietary# 67
Released
Oct 22, 2024
# 12
Knowledge Cutoff
Apr 24
# 8
Context Length
200K
Benchmarks
# 65
Code RankedAGI
46.6%
# 25
Aider Polyglot
51.6%
# 26
SWEBench Verified
49.0%
# 8
WebDev Arena
1245.4
# 23
LiveCodeBench v6
36.4%
# 30
LiveCodeBench v5
39.8%
# 24
Codeforces ELO
717
# 18
Code LMArena
1313
# 10
Code LiveBench (old)
67.1%
# 47
GPQA Diamond
65.0%
# 20
Reason LiveBench (old)
58.7%
# 24
ELO LMArena
1355
# 56
AIME 2025 I & II
3.0%
# 27
Math LiveBench (old)
51.3%
# 17
MATH
78.3%
# 24
MATH 500
78.3%
# 31
Humanity Last Exam
4.8%
# 1
Human Eval
93.7%
# 6
Human Eval+
86.2%
# 19
NYT Connections
17.7%
# 16
MMLU Pro
78.0%
# 17
MMLU
88.0%
# 21
MMMU
70.4%
# 26
Halluc. Hughes
4.6%
# 3
Aidan Bench
2691
# 45
AIME 2024
16.0%
# 31
IF LiveBench (old)
69.3%
# 20
Avg LiveBench (old)
60.7%
# 6
IF Evaluation
89.3%
Pricing
# 30
Input Cost /M
$3
# 35
Output Cost /M
$15
# 13
Cached Cost /M
$0.3