o3 mini high
LatestOpenAI
•
Proprietary# 32
Released
Jan 31, 2025
# 13
Knowledge Cutoff
Oct 23
# 6
Context Length
200K
Benchmarks
# 12
Code RankedAGI
67.4%
# 12
Aider Polyglot
60.4%
# 15
SWEBench Verified
49.3%
# 14
WebDev Arena
1147.27
# 6
LiveCodeBench v6
68.9%
# 1
LiveCodeBench v5
80.5%
# 3
Codeforces ELO
2130
# 15
Code LMArena
1332
# 2
Code LiveBench (old)
82.7%
# 12
GPQA Diamond
79.7%
# 3
Reason LiveBench (old)
89.6%
# 24
ELO LMArena
1355
# 12
AIME 2025 I & II
86.5%
# 5
Math LiveBench (old)
76.5%
# 1
MATH
97.9%
# 8
Humanity Last Exam
14.0%
# 3
NYT Connections
61.4%
# 15
MMLU
86.9%
# 2
Halluc. Hughes
0.8%
# 9
AIME 2024
87.3%
# 2
IF LiveBench (old)
84.4%
# 2
Avg LiveBench (old)
75.8%
Pricing
# 24
Input Cost /M
$1.1
# 27
Output Cost /M
$4.4
# 13
Cached Cost /M
$0.55