Claude 3.5 Sonnet
LatestAnthropic
•
Proprietary# 113
Released
Oct 22, 2024
# 16
Knowledge Cutoff
Apr 24
# 10
Context Length
200K
Benchmarks
# 135
Code RankedAGI
39.4%
# 48
SWEBench Verified
49.0%
# 155
Agentic RankedAGI
32.7%
# 34
LiveCodeBench v6
36.4%
# 27
LiveCodeBench v5
39.8%
# 12
Code LMArena
1313
# 27
Codeforces ELO
717
# 19
Aider Polyglot
51.6%
# 9
Code LiveBench (old)
67.1%
# 165
Reason RankedAGI
35.8%
# 68
HLE
4.8%
# 69
GPQA Diamond
65.0%
# 55
Text Arena
1355
# 56
AIME 2025 I & II
3.0%
# 37
AIME 2024
16.0%
# 1
Human Eval
93.7%
# 6
Human Eval+
86.2%
# 26
NYT Connections
17.7%
# 24
MMLU Pro
78.0%
# 14
MMLU
88.0%
# 21
MMMU
70.4%
# 27
Halluc. Hughes
4.6%
# 3
Aidan Bench
2691
# 17
Avg LiveBench (old)
60.7%
# 6
IF Evaluation
89.3%
# 28
Coding LiveBench 25.4
32.3%
# 28
Data LiveBench
52.8%
# 8
Language LiveBench
53.8%
# 5
Quality Artificial Analysis
80
# 203
Math RankedAGI
37.2%
# 155
RAGI RankedAGI
40.6%
Pricing
# 39
Input Cost /M
$3
# 46
Output Cost /M
$15
# 26
Cached Cost /M
$0.3