Claude Opus 5 vs Step 5 Preview

Change models
Claude Opus 5Anthropic
IndexAll categories, out of 100.
73.8 ±1.9 best
68.4 ±3.7
GapIndex points below the top.
Top
−5.4
CoverageIndex weight that has results.
100% best
90%
SpeedOutput tokens per second.
48/s
105/s best
EloLMArena rating and rank.
1502 best
1443
ReleasedMonth of release.
2026-07
2026-09
Input / 1MUS dollars per 1M input tokens.
$5.00
$1.00 best
Output / 1MUS dollars per 1M output tokens.
$25.00
$2.70 best
Batch output / 1MBatch price per 1M output tokens.
$12.50
—
1K chats2K tokens in, 500 out, each.
$22.50
$3.35 best
1K documents20K tokens in, 1K out, each.
$125.00
$22.70 best
1K agent tasks200K tokens in, 20K out, each.
$1,500.00
$254.00 best
ContextMaximum tokens in one request.
1M
1M
AgenticMulti-step tasks with tools.
71.6 best
68.8
CodingCode writing and repair.
74.8 best
71.4
ReasoningLogic problems and puzzles.
75.4 best
67.4
MultimodalTasks with images and text.
72.5 best
62.2
KnowledgeFacts and expert knowledge.
73.1 best
63.5
MultilingualTasks in many languages.
63.6
—
InstructionTasks with strict prompt rules.
55.6
—
MathMath problems.
74.8 best
61.7

Benchmarks

43 shared
Only Claude Opus 5: 107Only Step 5 Preview: 19
Claude Opus 5
Step 5 Preview
54.9% best7 / 222
46.5%23 / 222
79.3%52 / 217
88.3% best2 / 217
29.1% best13 / 214
20.9%26 / 214
60.9% best7 / 164
36.4%64 / 164
61.1% best7 / 133
54.3%24 / 133
88.4% best11 / 123
70.4%48 / 123
56.4%26 / 114
58.9% best13 / 114
84.7% best6 / 113
76.4%44 / 113
55.3% best2 / 90
40.4%35 / 90
47.1% best5 / 82
27.3%39 / 82
74.6% best40 / 80
63.5%69 / 80
91.2% best8 / 80
83.8%44 / 80
58.6% best10 / 79
50.7%42 / 79
57.5% best8 / 77
39.0%31 / 77
64.7% best4 / 77
46.5%26 / 77
EMBFinance
73.6% best6 / 75
57.1%41 / 75
BrowseCompAgentic
90.8% best8 / 63
88.7%12 / 63
53.5% best10 / 61
31.8%23 / 61
DeepSWEAgentic
68.8% best24 / 59
67.7%27 / 59
4.4 best17 / 54
−0.930 / 54