Inference
EN

Claude Fable 5.1 vs Muse Wonder Full

Change models
Claude Fable 5.1Anthropic
Muse Wonder FullMeta
IndexAll categories, out of 100.
76.2 ±2.4
77.3 ±5.3 best
GapIndex points below the top.
−1.1 · no clear gap
Top
CoverageIndex weight that has results.
95% best
80%
SpeedOutput tokens per second.
69/s
—
EloLMArena rating and rank.
1512
—
ReleasedMonth of release.
2026-09
2026-10
Input / 1MUS dollars per 1M input tokens.
$10.00
—
Output / 1MUS dollars per 1M output tokens.
$50.00
—
Batch output / 1MBatch price per 1M output tokens.
$25.00
—
1K chats2K tokens in, 500 out, each.
$45.00
—
1K documents20K tokens in, 1K out, each.
$250.00
—
1K agent tasks200K tokens in, 20K out, each.
$3,000.00
—
ContextMaximum tokens in one request.
1M
1M
AgenticMulti-step tasks with tools.
71.1
78.2 best
CodingCode writing and repair.
77.7 best
77.3
ReasoningLogic problems and puzzles.
75.8 best
75.6
MultimodalTasks with images and text.
77.9
—
KnowledgeFacts and expert knowledge.
77.2 best
71.7
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
74.3
—
MathMath problems.
77.0
79.6 best

Benchmarks

37 shared
Only Claude Fable 5.1: 95Only Muse Wonder Full: 5
Claude Fable 5.1
Muse Wonder Full
Artificial Analysis Humanity's Last ExamKnowledge
59.1% best3 / 222
53.6%9 / 222
Artificial Analysis Long Context ReasoningReasoning
85.3% best5 / 217
83.7%10 / 217
Critical Physics TasksReasoning
29.7%12 / 214
32.9% best2 / 214
Artificial Analysis Omniscience AccuracyKnowledge
68.4% best2 / 164
41.0%46 / 164
LegalBenchLegal
88.5% best2 / 149
87.4%4 / 149
GDPval-AA normalizedAgentic
62.7%6 / 133
67.9% best2 / 133
Vibe Code Bench v1.1Coding
90.3% best6 / 123
86.0%14 / 123
Artificial Analysis SciCodeCoding
63.1% best3 / 114
62.6%4 / 114
MedScribeHealthcare
91.3% best2 / 109
89.3%8 / 109
MedCodeHealthcare
53.5% best8 / 106
49.9%18 / 106
Legal Research BenchLegal
55.3% best2 / 90
53.4%6 / 90
Tax Agent BenchFinance
50.7% best2 / 82
43.0%8 / 82
Finance Agent (v2)Finance
58.9%8 / 79
59.1% best6 / 79
Code MigrationCoding
54.6%11 / 77
61.0% best7 / 77
EMBFinance
76.7% best1 / 75
71.9%10 / 75
ProgramBenchCoding
7.0% best3 / 65
0.5%19 / 65
Terminal-Bench 4.0Agentic
58.1%6 / 61
58.1%6 / 61
DeepSWEAgentic
67.4%29 / 59
79.0% best1 / 59
Agent Arena steerabilityAgentic
14.8 best5 / 54
9.612 / 54
Agent Arena task outcomeAgentic
14.4 best5 / 54
11.19 / 54