Inference
EN

Claude Opus 5.5 vs Muse Wonder Full

Change models
Claude Opus 5.5Anthropic
Muse Wonder FullMeta
IndexAll categories, out of 100.
78.6 ±2.4 best
77.3 ±5.3
GapIndex points below the top.
Top
−1.3 · no clear gap
CoverageIndex weight that has results.
95% best
80%
SpeedOutput tokens per second.
76/s
—
EloLMArena rating and rank.
1513
—
ReleasedMonth of release.
2026-09
2026-10
Input / 1MUS dollars per 1M input tokens.
$4.00
—
Output / 1MUS dollars per 1M output tokens.
$20.00
—
Batch output / 1MBatch price per 1M output tokens.
$10.00
—
1K chats2K tokens in, 500 out, each.
$18.00
—
1K documents20K tokens in, 1K out, each.
$100.00
—
1K agent tasks200K tokens in, 20K out, each.
$1,200.00
—
ContextMaximum tokens in one request.
1M
1M
AgenticMulti-step tasks with tools.
74.1
78.2 best
CodingCode writing and repair.
81.8 best
77.3
ReasoningLogic problems and puzzles.
80.8 best
75.6
MultimodalTasks with images and text.
74.1
—
KnowledgeFacts and expert knowledge.
79.2 best
71.7
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
59.6
—
MathMath problems.
77.7
79.6 best

Benchmarks

39 shared
Only Claude Opus 5.5: 118Only Muse Wonder Full: 3
Claude Opus 5.5
Muse Wonder Full
Artificial Analysis Humanity's Last ExamKnowledge
61.4% best2 / 222
53.6%9 / 222
Artificial Analysis Long Context ReasoningReasoning
84.7% best6 / 217
83.7%10 / 217
Critical Physics TasksReasoning
31.7%4 / 214
32.9% best2 / 214
Artificial Analysis Omniscience AccuracyKnowledge
65.5% best3 / 164
41.0%46 / 164
GDPval-AA normalizedAgentic
68.3% best1 / 133
67.9%2 / 133
Vibe Code Bench v1.1Coding
90.3% best6 / 123
86.0%14 / 123
Artificial Analysis SciCodeCoding
66.9% best2 / 114
62.6%4 / 114
MedScribeHealthcare
91.4% best1 / 109
89.3%8 / 109
MedCodeHealthcare
49.8%19 / 106
49.9% best18 / 106
Legal Research BenchLegal
50.5%7 / 90
53.4% best6 / 90
Tax Agent BenchFinance
44.5% best6 / 82
43.0%8 / 82
Finance Agent (v2)Finance
58.6%10 / 79
59.1% best6 / 79
Code MigrationCoding
66.6% best4 / 77
61.0%7 / 77
EMBFinance
75.9% best2 / 75
71.9%10 / 75
ProgramBenchCoding
18.5% best2 / 65
0.5%19 / 65
Terminal-Bench 4.0Agentic
65.2% best1 / 61
58.1%6 / 61
DeepSWEAgentic
74.2%10 / 59
79.0% best1 / 59
Agent Arena steerabilityAgentic
18.6 best3 / 54
9.612 / 54
Agent Arena task outcomeAgentic
20.7 best2 / 54
11.19 / 54
ProofBench v1.1Math
100.0% best1 / 54
94.0%11 / 54