Inference
EN

Muse Spark 1.3 vs Muse Wonder

Change models
Muse Spark 1.3Meta
Muse WonderMeta
IndexAll categories, out of 100.
70.1 ±4.0
73.1 ±4.0 best
GapIndex points below the top.
−3.0 · no clear gap
Top
CoverageIndex weight that has results.
85%
85%
SpeedOutput tokens per second.
76/s
80/s best
EloLMArena rating and rank.
1491
1503 best
ReleasedMonth of release.
2026-09
2026-10
Input / 1MUS dollars per 1M input tokens.
$1.25 best
$4.00
Output / 1MUS dollars per 1M output tokens.
$4.25 best
$20.00
1K chats2K tokens in, 500 out, each.
$4.63 best
$18.00
1K documents20K tokens in, 1K out, each.
$29.25 best
$100.00
1K agent tasks200K tokens in, 20K out, each.
$335.00 best
$1,200.00
ContextMaximum tokens in one request.
1.05M
1.05M
AgenticMulti-step tasks with tools.
70.4
73.0 best
CodingCode writing and repair.
68.8
72.3 best
ReasoningLogic problems and puzzles.
68.1
73.1 best
MultimodalTasks with images and text.
—
—
KnowledgeFacts and expert knowledge.
67.5
68.6 best
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
84.4 best
74.4
MathMath problems.
68.6
73.5 best

Benchmarks

41 shared
Only Muse Spark 1.3: 18Only Muse Wonder: 21
Muse Spark 1.3
Muse Wonder
Artificial Analysis Humanity's Last ExamKnowledge
48.7%15 / 222
50.1% best12 / 222
Artificial Analysis Long Context ReasoningReasoning
83.0% best18 / 217
80.7%37 / 217
Critical Physics TasksReasoning
24.9%21 / 214
28.3% best15 / 214
OTIS Mock AIME 2024-2025Math
99.2%19 / 200
99.4% best17 / 200
Artificial Analysis Omniscience AccuracyKnowledge
43.6% best41 / 164
38.0%59 / 164
GDPval-AA normalizedAgentic
59.0%10 / 133
63.5% best5 / 133
Vibe Code Bench v1.1Coding
82.9%23 / 123
83.1% best22 / 123
Artificial Analysis SciCodeCoding
58.8%14 / 114
60.3% best10 / 114
FrontierMath-Tiers-1-3-v2-PrivateMath
74.0%25 / 90
82.1% best18 / 90
Legal Research BenchLegal
40.9%33 / 90
50.5% best7 / 90
Tax Agent BenchFinance
37.9% best14 / 82
36.3%15 / 82
LiveBench Data AnalysisReasoning
79.6%13 / 80
81.6% best5 / 80
LiveBench ReasoningReasoning
89.7%15 / 80
90.6% best11 / 80
Mystery Game PuzzlesReasoning
25.0%37 / 80
51.0% best15 / 80
Finance Agent (v2)Finance
58.9% best8 / 79
56.6%18 / 79
Code MigrationCoding
27.6%44 / 77
62.6% best6 / 77
EMBFinance
62.7%31 / 75
67.1% best18 / 75
FrontierMath-Tier-4-v2-PrivateMath
46.3%24 / 72
70.7% best17 / 72
LiveBench Agentic CodingAgentic
64.1% best8 / 66
62.7%9 / 66
LiveBench CodingCoding
81.1% best15 / 66
80.7%16 / 66