o3
LatestOpenAI
•
Proprietary# 71
Released
Apr 16, 2025
# 14
Knowledge Cutoff
Jun 24
# 9
Context Length
200K
Benchmarks
# 71
Code RankedAGI
57.3%
# 36
SWEBench Verified
69.1%
# 93
Agentic RankedAGI
45.4%
# 20
Browse Comp
49.7%
# 20
LiveCodeBench v6
75.5%
# 3
Coding LiveBench 25.5
77.9%
# 2
Code LMArena
1460
# 6
Codeforces ELO
2706
# 5
Aider Polyglot
79.6%
# 71
Reason RankedAGI
56.2%
# 54
HLE
14.7%
# 31
HLE w/ Tools
24.3%
# 25
GPQA Diamond
87.7%
# 6
Reason LiveBench 25.5
91.0%
# 28
Text Arena
1431
# 21
AIME 2025 I & II
88.9%
# 8
AIME 2024
91.6%
# 11
Math LiveBench 25.5
80.7%
# 3
MMMU
82.9%
# 37
Halluc. Hughes
6.8%
# 3
IF LiveBench 25.5
84.3%
# 7
Avg LiveBench 25.5
72.4%
# 3
Coding LiveBench 25.4
77.9%
# 3
Reasoning LiveBench 25.4
91.0%
# 5
Agentic LiveBench 25.5
28.3%
# 13
MMMU Pro
76.4%
# 95
Math RankedAGI
63.3%
# 74
RAGI RankedAGI
53.1%
# 20
𝜏²-Bench Telecom
48.2%
# 7
𝜏²-Bench Retail
80.2%
# 1
𝜏²-Bench Airline
64.8%
# 4
HealthBench Hard
31.6%
# 55
GDPval AA
754
# 19
Svelte Bench v1
30.0%
# 62
Code DesignArena
1070
# 34
Toolathlon Pass@1
17.0%
Pricing
# 37
Input Cost /M
$2
# 42
Output Cost /M
$8
# 27
Cached Cost /M
$0.5