Inference
EN

Gemini 4 Carbon vs Muse Wonder Full

Change models
Gemini 4 CarbonGoogle
Muse Wonder FullMeta
IndexAll categories, out of 100.
82.7 ±5.1 best
77.3 ±5.3
GapIndex points below the top.
Top
−5.4 · no clear gap
CoverageIndex weight that has results.
80%
80%
SpeedOutput tokens per second.
31/s
—
EloLMArena rating and rank.
1538
—
ReleasedMonth of release.
2026-10
2026-10
Input / 1MUS dollars per 1M input tokens.
$2.00
—
Output / 1MUS dollars per 1M output tokens.
$10.00
—
Batch output / 1MBatch price per 1M output tokens.
$5.00
—
1K chats2K tokens in, 500 out, each.
$9.00
—
1K documents20K tokens in, 1K out, each.
$50.00
—
1K agent tasks200K tokens in, 20K out, each.
$600.00
—
ContextMaximum tokens in one request.
1.05M best
1M
AgenticMulti-step tasks with tools.
83.2 best
78.2
CodingCode writing and repair.
80.5 best
77.3
ReasoningLogic problems and puzzles.
83.9 best
75.6
MultimodalTasks with images and text.
—
—
KnowledgeFacts and expert knowledge.
79.5 best
71.7
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
—
—
MathMath problems.
77.2
79.6 best

Benchmarks

12 shared
Only Gemini 4 Carbon: 27Only Muse Wonder Full: 30
Gemini 4 Carbon
Muse Wonder Full
Artificial Analysis Long Context ReasoningReasoning
79.3%52 / 217
83.7% best10 / 217
Artificial Analysis Omniscience AccuracyKnowledge
58.3% best9 / 164
41.0%46 / 164
Vibe Code Bench v1.1Coding
89.1% best9 / 123
86.0%14 / 123
Legal Research BenchLegal
55.8% best1 / 90
53.4%6 / 90
Tax Agent BenchFinance
51.8% best1 / 82
43.0%8 / 82
Terminal-Bench 4.0Agentic
63.1% best4 / 61
58.1%6 / 61
Agent Arena steerabilityAgentic
19.1 best2 / 54
9.612 / 54
Agent Arena task outcomeAgentic
18.4 best4 / 54
11.19 / 54
HLAB-Vals-1.1Legal
17.5%8 / 35
23.3% best5 / 35
Artificial Analysis AutomationBenchAgentic
77.8% best1 / 23
45.4%21 / 23
Artificial Analysis GDP.pdfAgentic
28.4%7 / 22
32.6% best1 / 22
Artificial Analysis Terminal-Bench v4.0Agentic
65.2% best2 / 22
57.1%7 / 22