Claude Opus 4.5 vs o3

Change models
o3OpenAI
IndexAll categories, out of 100.
51.4 ±2.1 best
51.2 ±3.9
GapIndex points below the top.
Top
−0.2 · no clear gap
CoverageIndex weight that has results.
100% best
95%
SpeedOutput tokens per second.
37/s
115/s best
EloLMArena rating and rank.
1451 best
1410
ReleasedMonth of release.
2025-11
2025-04
Input / 1MUS dollars per 1M input tokens.
$5.00
$2.00 best
Output / 1MUS dollars per 1M output tokens.
$25.00
$8.00 best
Batch output / 1MBatch price per 1M output tokens.
$12.50
$4.00 best
1K chats2K tokens in, 500 out, each.
$22.50
$8.00 best
1K documents20K tokens in, 1K out, each.
$125.00
$48.00 best
1K agent tasks200K tokens in, 20K out, each.
$1,500.00
$560.00 best
ContextMaximum tokens in one request.
200K
200K
AgenticMulti-step tasks with tools.
54.0 best
52.3
CodingCode writing and repair.
54.3 best
50.5
ReasoningLogic problems and puzzles.
51.2
52.1 best
MultimodalTasks with images and text.
41.7
52.6 best
KnowledgeFacts and expert knowledge.
54.0 best
53.1
MultilingualTasks in many languages.
55.8
—
InstructionTasks with strict prompt rules.
41.0
55.0 best
MathMath problems.
50.6 best
48.6

Benchmarks

33 shared
Only Claude Opus 4.5: 69Only o3: 13
Claude Opus 4.5
o3
85.5% best73 / 220
81.8%85 / 220
13.2%127 / 216
20.1% best108 / 216
70.7%105 / 211
74.7% best92 / 211
0.3%132 / 207
1.1% best113 / 207
81.7%86 / 203
83.9% best81 / 203
81.0%92 / 193
82.7% best87 / 193
82.8%65 / 149
83.8% best47 / 149
TaxEval v2Finance
74.3%38 / 143
74.6% best34 / 143
86.3% best48 / 143
80.7%64 / 143
75.0%80 / 142
83.9% best44 / 142
43.0%89 / 140
71.4% best37 / 140
79.5%69 / 135
84.1% best52 / 135
MMLU ProKnowledge
85.6%55 / 135
85.6%55 / 135
CorpFin v2Finance
61.3% best54 / 131
59.7%74 / 131
MedScribeHealthcare
83.2% best49 / 111
76.7%76 / 111
71.2% best69 / 109
70.1%71 / 109
MedCodeHealthcare
45.2%38 / 108
47.3% best34 / 108
68.7% best11 / 97
65.7%40 / 97
AIMEMath
76.9%50 / 95
85.3% best37 / 95
MedQAHealthcare
93.2%24 / 95
96.1% best5 / 95