Inference
EN

MiMo-V2.6-Pro vs Muse Wonder Full

Change models
MiMo-V2.6-ProXiaomi
Muse Wonder FullMeta
IndexAll categories, out of 100.
70.2 ±4.7
77.3 ±5.3 best
GapIndex points below the top.
−7.1
Top
CoverageIndex weight that has results.
80%
80%
SpeedOutput tokens per second.
31/s
—
EloLMArena rating and rank.
1492
—
ReleasedMonth of release.
2026-09
2026-10
Input / 1MUS dollars per 1M input tokens.
$0.435
—
Output / 1MUS dollars per 1M output tokens.
$0.87
—
1K chats2K tokens in, 500 out, each.
$1.31
—
1K documents20K tokens in, 1K out, each.
$9.57
—
1K agent tasks200K tokens in, 20K out, each.
$104.40
—
ContextMaximum tokens in one request.
1.05M best
1M
AgenticMulti-step tasks with tools.
72.1
78.2 best
CodingCode writing and repair.
66.6
77.3 best
ReasoningLogic problems and puzzles.
68.3
75.6 best
MultimodalTasks with images and text.
—
—
KnowledgeFacts and expert knowledge.
64.2
71.7 best
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
—
—
MathMath problems.
71.3
79.6 best

Benchmarks

34 shared
Only MiMo-V2.6-Pro: 24Only Muse Wonder Full: 8
MiMo-V2.6-Pro
Muse Wonder Full
Artificial Analysis Humanity's Last ExamKnowledge
49.4%14 / 222
53.6% best9 / 222
Artificial Analysis Long Context ReasoningReasoning
86.3% best3 / 217
83.7%10 / 217
Critical Physics TasksReasoning
26.6%19 / 214
32.9% best2 / 214
Artificial Analysis Omniscience AccuracyKnowledge
34.8%67 / 164
41.0% best46 / 164
GDPval-AA normalizedAgentic
59.2%9 / 133
67.9% best2 / 133
Vibe Code Bench v1.1Coding
85.2%16 / 123
86.0% best14 / 123
Artificial Analysis SciCodeCoding
60.9%8 / 114
62.6% best4 / 114
MedScribeHealthcare
88.3%13 / 109
89.3% best8 / 109
MedCodeHealthcare
45.0%39 / 106
49.9% best18 / 106
Legal Research BenchLegal
47.1%16 / 90
53.4% best6 / 90
Tax Agent BenchFinance
32.4%22 / 82
43.0% best8 / 82
Finance Agent (v2)Finance
57.3%16 / 79
59.1% best6 / 79
Code MigrationCoding
43.0%26 / 77
61.0% best7 / 77
EMBFinance
62.9%30 / 75
71.9% best10 / 75
ProgramBenchCoding
0.5%19 / 65
0.5%19 / 65
Terminal-Bench 4.0Agentic
31.3%24 / 61
58.1% best6 / 61
DeepSWEAgentic
71.9%18 / 59
79.0% best1 / 59
Agent Arena steerabilityAgentic
1.525 / 54
9.6 best12 / 54
Agent Arena task outcomeAgentic
5.016 / 54
11.1 best9 / 54
ProofBench v1.1Math
70.0%21 / 54
94.0% best11 / 54