o3 vs o4-mini

Change models
o3OpenAI
o4-miniOpenAI
IndexAll categories, out of 100.
51.8 ±3.9 best
44.7 ±3.0
GapIndex points below the top.
Top
−7.1
CoverageIndex weight that has results.
95%
100% best
SpeedOutput tokens per second.
65/s
122/s best
EloLMArena rating and rank.
1410 best
1353
ReleasedMonth of release.
2025-04
2025-04
Input / 1MUS dollars per 1M input tokens.
$2.00
$1.10 best
Output / 1MUS dollars per 1M output tokens.
$8.00
$4.40 best
Batch output / 1MBatch price per 1M output tokens.
$4.00
—
1K chats2K tokens in, 500 out, each.
$8.00
$4.40 best
1K documents20K tokens in, 1K out, each.
$48.00
$26.40 best
1K agent tasks200K tokens in, 20K out, each.
$560.00
$308.00 best
ContextMaximum tokens in one request.
200K
200K
AgenticMulti-step tasks with tools.
51.8 best
38.5
CodingCode writing and repair.
51.0 best
45.2
ReasoningLogic problems and puzzles.
52.9 best
44.9
MultimodalTasks with images and text.
53.1 best
52.0
KnowledgeFacts and expert knowledge.
53.7 best
40.8
MultilingualTasks in many languages.
—
35.5
InstructionTasks with strict prompt rules.
55.7 best
53.2
MathMath problems.
49.1
50.5 best

Benchmarks

44 shared
Only o3: 2Only o4-mini: 12
o3
o4-mini
20.1% best112 / 222
16.5%122 / 222
74.7% best97 / 217
61.0%141 / 217
81.8% best82 / 217
79.6%91 / 217
1.1% best118 / 214
0.6%132 / 214
83.9% best78 / 200
81.7%83 / 200
82.7% best91 / 199
78.4%103 / 199
83.8% best47 / 149
79.2%97 / 149
80.7% best67 / 149
55.6%89 / 149
71.4% best38 / 146
68.7%51 / 146
TaxEval v2Finance
74.6%34 / 143
74.8% best29 / 143
83.9% best44 / 142
82.2%53 / 142
84.1% best52 / 135
74.5%83 / 135
MMLU ProKnowledge
85.6% best55 / 135
80.6%86 / 135
6.5% best72 / 131
6.1%75 / 131
CorpFin v2Finance
59.7% best74 / 131
59.0%79 / 131
60.8% best67 / 126
58.7%69 / 126
2.9%80 / 122
7.5% best62 / 122
64.3%63 / 116
68.0% best60 / 116
70.1% best74 / 113
69.2%77 / 113
MedScribeHealthcare
76.7% best74 / 109
69.1%97 / 109