MiMo-V2.5 vs Qwen3.6 Plus

Change models
MiMo-V2.5Xiaomi
IndexAll categories, out of 100.
54.8 ±6.8 best
54.6 ±2.1
GapIndex points below the top.
Top
−0.2 · no clear gap
CoverageIndex weight that has results.
75%
100% best
SpeedOutput tokens per second.
4/s
44/s best
EloLMArena rating and rank.
1428
1436 best
ReleasedMonth of release.
2026-04
2026-04
Input / 1MUS dollars per 1M input tokens.
$0.40
$0.325 best
Output / 1MUS dollars per 1M output tokens.
$2.00
$1.95 best
1K chats2K tokens in, 500 out, each.
$1.80
$1.63 best
1K documents20K tokens in, 1K out, each.
$10.00
$8.45 best
1K agent tasks200K tokens in, 20K out, each.
$120.00
$104.00 best
ContextMaximum tokens in one request.
1M
1M
AgenticMulti-step tasks with tools.
56.5 best
55.0
CodingCode writing and repair.
55.5
56.2 best
ReasoningLogic problems and puzzles.
—
51.2
MultimodalTasks with images and text.
57.2 best
56.1
KnowledgeFacts and expert knowledge.
50.5
54.1 best
MultilingualTasks in many languages.
—
54.1
InstructionTasks with strict prompt rules.
—
52.4
MathMath problems.
52.7
54.9 best

Benchmarks

25 shared
Only MiMo-V2.5: 3Only Qwen3.6 Plus: 66
MiMo-V2.5
Qwen3.6 Plus
78.9%101 / 149
84.2% best37 / 149
TaxEval v2Finance
71.8%72 / 143
74.7% best31 / 143
81.5%60 / 142
86.0% best24 / 142
81.6%61 / 135
87.4% best38 / 135
MMLU ProKnowledge
82.9%75 / 135
87.7% best24 / 135
CorpFin v2Finance
59.9%71 / 131
61.9% best49 / 131
42.2% best75 / 123
25.6%84 / 123
MedScribeHealthcare
72.2%90 / 109
77.0% best73 / 109
MedCodeHealthcare
31.9%91 / 106
36.9% best76 / 106
59.3%67 / 97
68.0% best20 / 97
MMMU ProMultimodal
80.0%45 / 92
84.2% best27 / 92
9.1%82 / 90
14.9% best75 / 90
SWE-benchCoding
71.0%60 / 87
73.4% best52 / 87
56.1%51 / 83
56.6% best47 / 83
7.8% best71 / 82
7.7%72 / 82
36.7%67 / 79
40.8% best63 / 79
14.2% best57 / 77
11.1%64 / 77
EMBFinance
55.1% best50 / 75
32.9%63 / 75
60.7% best40 / 75
53.2%56 / 75
77.9%28 / 53
78.8% best21 / 53