Inference
EN

Claude Fable 5.2 vs Muse Spark 1.3

Change models
Claude Fable 5.2AnthropicSmartest
Muse Spark 1.3Meta
IndexAll categories, out of 100.
85.6 ±3.7 best
70.1 ±4.0
GapIndex points below the top.
Top
−15.5
CoverageIndex weight that has results.
90% best
85%
SpeedOutput tokens per second.
64/s
76/s best
EloLMArena rating and rank.
—
1491
ReleasedMonth of release.
2026-11
2026-09
Input / 1MUS dollars per 1M input tokens.
$8.00
$1.25 best
Output / 1MUS dollars per 1M output tokens.
$40.00
$4.25 best
Batch output / 1MBatch price per 1M output tokens.
$20.00
—
1K chats2K tokens in, 500 out, each.
$36.00
$4.63 best
1K documents20K tokens in, 1K out, each.
$200.00
$29.25 best
1K agent tasks200K tokens in, 20K out, each.
$2,400.00
$335.00 best
ContextMaximum tokens in one request.
1M
1.05M best
AgenticMulti-step tasks with tools.
82.2 best
70.4
CodingCode writing and repair.
86.2 best
68.8
ReasoningLogic problems and puzzles.
86.5 best
68.1
MultimodalTasks with images and text.
82.0
—
KnowledgeFacts and expert knowledge.
87.1 best
67.5
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
—
84.4
MathMath problems.
79.3 best
68.6

Benchmarks

24 shared
Only Claude Fable 5.2: 47Only Muse Spark 1.3: 35
Claude Fable 5.2
Muse Spark 1.3
Artificial Analysis Humanity's Last ExamKnowledge
74.3% best1 / 222
48.7%15 / 222
Artificial Analysis Long Context ReasoningReasoning
83.3% best13 / 217
83.0%18 / 217
Critical Physics TasksReasoning
35.4% best1 / 214
24.9%21 / 214
OTIS Mock AIME 2024-2025Math
99.7% best14 / 200
99.2%19 / 200
Artificial Analysis Omniscience AccuracyKnowledge
79.1% best1 / 164
43.6%41 / 164
Vibe Code Bench v1.1Coding
93.8% best1 / 123
82.9%23 / 123
Artificial Analysis SciCodeCoding
67.7% best1 / 114
58.8%14 / 114
Chess PuzzlesReasoning
79.0% best1 / 90
37.0%39 / 90
FrontierMath-Tiers-1-3-v2-PrivateMath
95.1% best1 / 90
74.0%25 / 90
Legal Research BenchLegal
46.2% best19 / 90
40.9%33 / 90
Tax Agent BenchFinance
32.1%23 / 82
37.9% best14 / 82
Mystery Game PuzzlesReasoning
90.0% best1 / 80
25.0%37 / 80
FrontierMath-Tier-4-v2-PrivateMath
100.0% best1 / 72
46.3%24 / 72
ProgramBenchCoding
19.0% best1 / 65
0.5%19 / 65
Terminal-Bench 4.0Agentic
65.2% best1 / 61
10.6%45 / 61
DeepSWEAgentic
74.3%9 / 59
75.4% best6 / 59
Agent Arena steerabilityAgentic
33.0 best1 / 54
2.622 / 54
Agent Arena task outcomeAgentic
22.0 best1 / 54
3.720 / 54
ProofBench v1.1Math
100.0% best1 / 54
55.0%30 / 54
Terminal-Bench-ScienceScience
55.7% best2 / 44
4.3%26 / 44