Inference
EN

Claude Sonnet 5 vs MiMo-V2.6-Flash

Change models
Claude Sonnet 5Anthropic
MiMo-V2.6-FlashXiaomi
IndexAll categories, out of 100.
64.2 ±2.7
64.3 ±4.4 best
GapIndex points below the top.
−0.1 · no clear gap
Top
CoverageIndex weight that has results.
95% best
90%
SpeedOutput tokens per second.
75/s best
28/s
EloLMArena rating and rank.
1442
1450 best
ReleasedMonth of release.
2026-06
2026-09
Input / 1MUS dollars per 1M input tokens.
$2.00
$0.14 best
Output / 1MUS dollars per 1M output tokens.
$10.00
$0.28 best
Batch output / 1MBatch price per 1M output tokens.
$5.00
—
1K chats2K tokens in, 500 out, each.
$9.00
$0.42 best
1K documents20K tokens in, 1K out, each.
$50.00
$3.08 best
1K agent tasks200K tokens in, 20K out, each.
$600.00
$33.60 best
ContextMaximum tokens in one request.
1M
1.05M best
AgenticMulti-step tasks with tools.
65.3
69.7 best
CodingCode writing and repair.
65.3 best
64.1
ReasoningLogic problems and puzzles.
66.5 best
59.2
MultimodalTasks with images and text.
59.6 best
50.3
KnowledgeFacts and expert knowledge.
59.0 best
52.3
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
55.8
—
MathMath problems.
64.8
68.9 best

Benchmarks

29 shared
Only Claude Sonnet 5: 57Only MiMo-V2.6-Flash: 12
Claude Sonnet 5
MiMo-V2.6-Flash
Artificial Analysis Humanity's Last ExamKnowledge
41.3% best41 / 222
35.1%62 / 222
Artificial Analysis Long Context ReasoningReasoning
82.0% best29 / 217
74.3%98 / 217
Critical Physics TasksReasoning
16.9% best43 / 214
12.0%57 / 214
Artificial Analysis Omniscience AccuracyKnowledge
40.1% best51 / 164
27.2%91 / 164
GDPval-AA normalizedAgentic
48.2%29 / 133
55.4% best19 / 133
Vibe Code Bench v1.1Coding
81.3% best29 / 123
79.0%33 / 123
Artificial Analysis SciCodeCoding
54.3% best34 / 114
51.3%46 / 114
Artificial Analysis MMMU-ProMultimodal
77.3% best40 / 113
73.1%66 / 113
MedScribeHealthcare
76.1%77 / 109
85.3% best29 / 109
MedCodeHealthcare
47.5% best33 / 106
41.1%58 / 106
Legal Research BenchLegal
41.8% best28 / 90
38.0%43 / 90
Terminal-Bench 2.1 · BenchLMAgentic
80.4%34 / 86
87.6% best13 / 86
Tax Agent BenchFinance
29.4% best36 / 82
28.5%37 / 82
Finance Agent (v2)Finance
53.9%29 / 79
56.3% best19 / 79
Code MigrationCoding
44.4% best23 / 77
40.9%29 / 77
EMBFinance
66.3% best22 / 75
65.5%24 / 75
Terminal-Bench 2.1 · Vals AIAgentic
74.5%18 / 75
76.4% best16 / 75
ProgramBenchCoding
0.0%34 / 65
0.5% best19 / 65
Terminal-Bench 4.0Agentic
10.6%45 / 61
24.2% best30 / 61
Agent Arena steerabilityAgentic
2.820 / 54
2.820 / 54