o4 mini
LatestOpenAI
•
Proprietary# 71
Released
Apr 16, 2025
# 14
Knowledge Cutoff
Jun 24
# 9
Context Length
200K
Benchmarks
# 83
Code RankedAGI
53.2%
# 37
SWEBench Verified
68.1%
# 107
Agentic RankedAGI
41.8%
# 7
Coding LiveBench 25.5
74.2%
# 13
Code LMArena
1334
# 21
Cyber Gym
2.5%
# 5
Codeforces ELO
2719
# 81
Reason RankedAGI
53.2%
# 49
HLE
18.1%
# 45
GPQA Diamond
81.4%
# 11
Reason LiveBench 25.5
78.5%
# 46
Text Arena
1390
# 32
AIME 2025 I & II
84.0%
# 10
Math LiveBench 25.5
81.0%
# 6
MMMU
81.6%
# 28
Halluc. Hughes
4.6%
# 6
IF LiveBench 25.5
81.8%
# 10
Avg LiveBench 25.5
67.4%
# 14
Coding LiveBench 25.4
61.8%
# 8
Agentic LiveBench 25.5
21.7%
# 105
Math RankedAGI
61.4%
# 87
RAGI RankedAGI
50.6%
# 28
Svelte Bench v1
13.3%
# 68
Code DesignArena
1026
# 35
Toolathlon Pass@1
14.8%
Pricing
# 31
Input Cost /M
$1.1
# 36
Output Cost /M
$4.4
# 25
Cached Cost /M
$0.28