Claude Opus 4.1 vs Claude Sonnet 4

Change models
IndexAll categories, out of 100.
47.1 ±5.0 best
42.9 ±5.2
GapIndex points below the top.
Top
−4.2 · no clear gap
CoverageIndex weight that has results.
90%
90%
SpeedOutput tokens per second.
14/s
49/s best
EloLMArena rating and rank.
1418 best
1339
ReleasedMonth of release.
2025-08
2025-05
Input / 1MUS dollars per 1M input tokens.
$15.00
$3.00 best
Output / 1MUS dollars per 1M output tokens.
$75.00
$15.00 best
Batch output / 1MBatch price per 1M output tokens.
$37.50
—
1K chats2K tokens in, 500 out, each.
$67.50
$13.50 best
1K documents20K tokens in, 1K out, each.
$375.00
$75.00 best
1K agent tasks200K tokens in, 20K out, each.
$4,500.00
$900.00 best
ContextMaximum tokens in one request.
200K
200K
AgenticMulti-step tasks with tools.
49.3
50.0 best
CodingCode writing and repair.
47.8 best
45.8
ReasoningLogic problems and puzzles.
54.6 best
39.0
MultimodalTasks with images and text.
50.6 best
48.9
KnowledgeFacts and expert knowledge.
49.1 best
44.2
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
—
—
MathMath problems.
41.5 best
38.6

Benchmarks

21 shared
Only Claude Opus 4.1: 5Only Claude Sonnet 4: 9
Claude Opus 4.1
Claude Sonnet 4
76.8%102 / 220
77.0% best100 / 220
64.4% best112 / 203
61.1%114 / 203
83.5% best54 / 149
83.0%62 / 149
TaxEval v2Finance
71.5% best78 / 143
69.6%97 / 143
64.6% best101 / 142
59.7%107 / 142
70.0% best94 / 135
69.4%95 / 135
MMLU ProKnowledge
87.2% best32 / 135
79.4%95 / 135
MedScribeHealthcare
71.8%94 / 111
72.4% best90 / 111
MedCodeHealthcare
41.4% best54 / 108
33.9%83 / 108
61.1%63 / 97
62.5% best54 / 97
AIMEMath
44.2% best65 / 95
38.5%71 / 95
MedQAHealthcare
92.5% best30 / 95
90.3%46 / 95
MMMU ProMultimodal
73.7% best58 / 92
72.4%64 / 92
MGSMMultilingual
94.2% best5 / 74
93.0%11 / 74
6.6% best41 / 69
4.1%49 / 69
0.0%33 / 64
0.0%33 / 64
IOI v1Coding
12.5% best32 / 61
6.5%41 / 61
93.0% best17 / 59
90.3%26 / 59
4.2% best26 / 55
0.0%45 / 55
82.4% best5 / 23
80.5%10 / 23