Inference
EN

Muse Wonder vs Muse Wonder Full

Change models
Muse WonderMeta
Muse Wonder FullMeta
IndexAll categories, out of 100.
73.1 ±4.0
77.3 ±5.3 best
GapIndex points below the top.
−4.2 · no clear gap
Top
CoverageIndex weight that has results.
85% best
80%
SpeedOutput tokens per second.
80/s
—
EloLMArena rating and rank.
1503
—
ReleasedMonth of release.
2026-10
2026-10
Input / 1MUS dollars per 1M input tokens.
$4.00
—
Output / 1MUS dollars per 1M output tokens.
$20.00
—
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.
1.05M best
1M
AgenticMulti-step tasks with tools.
73.0
78.2 best
CodingCode writing and repair.
72.3
77.3 best
ReasoningLogic problems and puzzles.
73.1
75.6 best
MultimodalTasks with images and text.
—
—
KnowledgeFacts and expert knowledge.
68.6
71.7 best
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
74.4
—
MathMath problems.
73.5
79.6 best

Benchmarks

37 shared
Only Muse Wonder: 25Only Muse Wonder Full: 5
Muse Wonder
Muse Wonder Full
Artificial Analysis Humanity's Last ExamKnowledge
50.1%12 / 222
53.6% best9 / 222
Artificial Analysis Long Context ReasoningReasoning
80.7%37 / 217
83.7% best10 / 217
Critical Physics TasksReasoning
28.3%15 / 214
32.9% best2 / 214
Artificial Analysis Omniscience AccuracyKnowledge
38.0%59 / 164
41.0% best46 / 164
GDPval-AA normalizedAgentic
63.5%5 / 133
67.9% best2 / 133
Vibe Code Bench v1.1Coding
83.1%22 / 123
86.0% best14 / 123
Artificial Analysis SciCodeCoding
60.3%10 / 114
62.6% best4 / 114
Legal Research BenchLegal
50.5%7 / 90
53.4% best6 / 90
Tax Agent BenchFinance
36.3%15 / 82
43.0% best8 / 82
Finance Agent (v2)Finance
56.6%18 / 79
59.1% best6 / 79
Code MigrationCoding
62.6% best6 / 77
61.0%7 / 77
EMBFinance
67.1%18 / 75
71.9% best10 / 75
Terminal-Bench 4.0Agentic
49.5%11 / 61
58.1% best6 / 61
DeepSWEAgentic
78.8%2 / 59
79.0% best1 / 59
Agent Arena steerabilityAgentic
2.622 / 54
9.6 best12 / 54
Agent Arena task outcomeAgentic
5.413 / 54
11.1 best9 / 54
ProofBench v1.1Math
84.0%15 / 54
94.0% best11 / 54
Public Benefits Bench v1.1Public benefits
70.2%6 / 50
73.7% best3 / 50
CyberBench v1.1Other
80.4% best1 / 48
73.6%11 / 48
MazeBench: visual spatial reasoning in a 3D open worldAgentic
0.0%23 / 48
17.0% best2 / 48