Inference
EN

Gemini 3.8 Flash vs Muse Spark 1.3

Change models
Gemini 3.8 FlashGoogle
Muse Spark 1.3Meta
IndexAll categories, out of 100.
67.8 ±2.5
70.1 ±4.0 best
GapIndex points below the top.
−2.3 · no clear gap
Top
CoverageIndex weight that has results.
95% best
85%
SpeedOutput tokens per second.
68/s
76/s best
EloLMArena rating and rank.
1500 best
1491
ReleasedMonth of release.
2026-09
2026-09
Input / 1MUS dollars per 1M input tokens.
$0.825 best
$1.25
Output / 1MUS dollars per 1M output tokens.
$4.13 best
$4.25
Batch output / 1MBatch price per 1M output tokens.
$1.88
—
1K chats2K tokens in, 500 out, each.
$3.71 best
$4.63
1K documents20K tokens in, 1K out, each.
$20.63 best
$29.25
1K agent tasks200K tokens in, 20K out, each.
$247.50 best
$335.00
ContextMaximum tokens in one request.
1.05M
1.05M
AgenticMulti-step tasks with tools.
65.3
70.4 best
CodingCode writing and repair.
66.2
68.8 best
ReasoningLogic problems and puzzles.
68.3 best
68.1
MultimodalTasks with images and text.
67.1
—
KnowledgeFacts and expert knowledge.
70.9 best
67.5
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
91.3 best
84.4
MathMath problems.
64.3
68.6 best

Benchmarks

52 shared
Only Gemini 3.8 Flash: 54Only Muse Spark 1.3: 7
Gemini 3.8 Flash
Muse Spark 1.3
Artificial Analysis Humanity's Last ExamKnowledge
47.8%20 / 222
48.7% best15 / 222
Artificial Analysis Long Context ReasoningReasoning
81.3%34 / 217
83.0% best18 / 217
Critical Physics TasksReasoning
18.3%34 / 214
24.9% best21 / 214
OTIS Mock AIME 2024-2025Math
98.9%22 / 200
99.2% best19 / 200
Artificial Analysis GPQA DiamondKnowledge
95.3% best2 / 199
93.5%8 / 199
Artificial Analysis Omniscience AccuracyKnowledge
50.5% best19 / 164
43.6%41 / 164
GDPval-AA normalizedAgentic
46.8%34 / 133
59.0% best10 / 133
Vibe Code Bench v1.1Coding
78.7%34 / 123
82.9% best23 / 123
Artificial Analysis SciCodeCoding
56.6%23 / 114
58.8% best14 / 114
Chess PuzzlesReasoning
58.0% best14 / 90
37.0%39 / 90
FrontierMath-Tiers-1-3-v2-PrivateMath
68.4%32 / 90
74.0% best25 / 90
Legal Research BenchLegal
38.9%38 / 90
40.9% best33 / 90
Terminal-Bench 2.1 · BenchLMAgentic
89.4% best5 / 86
88.8%7 / 86
Tax Agent BenchFinance
32.1%23 / 82
37.9% best14 / 82
LiveBench Data AnalysisReasoning
54.0%76 / 80
79.6% best13 / 80
LiveBench ReasoningReasoning
89.3%18 / 80
89.7% best15 / 80
Mystery Game PuzzlesReasoning
47.0% best16 / 80
25.0%37 / 80
Finance Agent (v2)Finance
61.4% best2 / 79
58.9%8 / 79
Code MigrationCoding
36.5% best35 / 77
27.6%44 / 77
EMBFinance
72.2% best8 / 75
62.7%31 / 75