Inkling-Small vs MiMo-V2.5

Change models
Inkling-SmallThinking Machines Lab
MiMo-V2.5Xiaomi
IndexAll categories, out of 100.
54.8 ±3.2 best
54.5 ±6.8
GapIndex points below the top.
Top
−0.3 · no clear gap
CoverageIndex weight that has results.
95% best
75%
SpeedOutput tokens per second.
99/s best
48/s
EloLMArena rating and rank.
1414
1427 best
ReleasedMonth of release.
2026-07
2026-04
Input / 1MUS dollars per 1M input tokens.
$0.45
$0.40 best
Output / 1MUS dollars per 1M output tokens.
$1.20 best
$2.00
1K chats2K tokens in, 500 out, each.
$1.50 best
$1.80
1K documents20K tokens in, 1K out, each.
$10.20
$10.00 best
1K agent tasks200K tokens in, 20K out, each.
$114.00 best
$120.00
ContextMaximum tokens in one request.
524K
1M best
AgenticMulti-step tasks with tools.
55.1
56.2 best
CodingCode writing and repair.
54.7
55.3 best
ReasoningLogic problems and puzzles.
55.5
—
MultimodalTasks with images and text.
54.5
57.0 best
KnowledgeFacts and expert knowledge.
51.8 best
50.3
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
65.6
—
MathMath problems.
55.0 best
52.3

Benchmarks

21 shared
Only Inkling-Small: 44Only MiMo-V2.5: 7
Inkling-Small
MiMo-V2.5
83.0% best62 / 149
78.9%101 / 149
TaxEval v2Finance
75.5% best18 / 143
71.8%72 / 143
85.9% best25 / 142
81.5%60 / 142
83.6% best55 / 135
81.6%61 / 135
MMLU ProKnowledge
85.6% best55 / 135
82.9%75 / 135
CorpFin v2Finance
69.6% best6 / 131
59.9%71 / 131
19.1%97 / 125
42.2% best77 / 125
MedScribeHealthcare
84.1% best38 / 109
72.2%90 / 109
MedCodeHealthcare
37.9% best74 / 106
31.9%91 / 106
62.3% best55 / 97
59.3%67 / 97
25.5% best65 / 92
9.1%84 / 92
SWE-benchCoding
82.2% best20 / 87
71.0%60 / 87
15.4% best62 / 84
7.8%73 / 84
55.9%52 / 83
56.1% best51 / 83
41.3% best62 / 79
36.7%67 / 79
13.7%59 / 77
14.2% best57 / 77
EMBFinance
31.8%64 / 75
55.1% best50 / 75
55.1%49 / 75
60.7% best40 / 75
74.0%40 / 54
77.9% best28 / 54
6.0%51 / 54
16.0% best46 / 54