Inference
EN

Gemini 4 Carbon vs Muse Wonder

Change models
Gemini 4 CarbonGoogle
Muse WonderMeta
IndexAll categories, out of 100.
82.7 ±5.1 best
73.1 ±4.0
GapIndex points below the top.
Top
−9.6
CoverageIndex weight that has results.
80%
85% best
SpeedOutput tokens per second.
31/s
80/s best
EloLMArena rating and rank.
1538 best
1503
ReleasedMonth of release.
2026-10
2026-10
Input / 1MUS dollars per 1M input tokens.
$2.00 best
$4.00
Output / 1MUS dollars per 1M output tokens.
$10.00 best
$20.00
Batch output / 1MBatch price per 1M output tokens.
$5.00
—
1K chats2K tokens in, 500 out, each.
$9.00 best
$18.00
1K documents20K tokens in, 1K out, each.
$50.00 best
$100.00
1K agent tasks200K tokens in, 20K out, each.
$600.00 best
$1,200.00
ContextMaximum tokens in one request.
1.05M
1.05M
AgenticMulti-step tasks with tools.
83.2 best
73.0
CodingCode writing and repair.
80.5 best
72.3
ReasoningLogic problems and puzzles.
83.9 best
73.1
MultimodalTasks with images and text.
—
—
KnowledgeFacts and expert knowledge.
79.5 best
68.6
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
—
74.4
MathMath problems.
77.2 best
73.5

Benchmarks

21 shared
Only Gemini 4 Carbon: 18Only Muse Wonder: 41
Gemini 4 Carbon
Muse Wonder
Artificial Analysis Long Context ReasoningReasoning
79.3%52 / 217
80.7% best37 / 217
OTIS Mock AIME 2024-2025Math
100.0% best1 / 200
99.4%17 / 200
Artificial Analysis Omniscience AccuracyKnowledge
58.3% best9 / 164
38.0%59 / 164
Vibe Code Bench v1.1Coding
89.1% best9 / 123
83.1%22 / 123
FrontierMath-Tiers-1-3-v2-PrivateMath
90.5% best6 / 90
82.1%18 / 90
Legal Research BenchLegal
55.8% best1 / 90
50.5%7 / 90
Tax Agent BenchFinance
51.8% best1 / 82
36.3%15 / 82
Mystery Game PuzzlesReasoning
73.0% best5 / 80
51.0%15 / 80
FrontierMath-Tier-4-v2-PrivateMath
95.1% best4 / 72
70.7%17 / 72
Terminal-Bench 4.0Agentic
63.1% best4 / 61
49.5%11 / 61
Agent Arena steerabilityAgentic
19.1 best2 / 54
2.622 / 54
Agent Arena task outcomeAgentic
18.4 best4 / 54
5.413 / 54
Furniture AssemblyReasoning
77.5% best6 / 37
54.2%13 / 37
HLAB-Vals-1.1Legal
17.5%8 / 35
23.8% best4 / 35
EBR-benchReasoning
87.6% best2 / 31
31.4%15 / 31
FrontierCode 2.0Coding
40.0% best5 / 27
24.3%15 / 27
Artificial Analysis AutomationBenchAgentic
77.8% best1 / 23
39.1%22 / 23
Artificial Analysis GDP.pdfAgentic
28.4%7 / 22
30.6% best5 / 22
Artificial Analysis Terminal-Bench v4.0Agentic
65.2% best2 / 22
52.0%10 / 22
cursorBench40Coding
60.4% best1 / 21
50.4%7 / 21