Inkling 2 Mini vs Inkling 2 Small

Change models
Inkling 2 MiniThinking Machines Lab
Inkling 2 SmallThinking Machines Lab
IndexAll categories, out of 100.
57.7 ±4.2
65.5 ±3.7 best
GapIndex points below the top.
−7.8
Top
CoverageIndex weight that has results.
90%
90%
SpeedOutput tokens per second.
88/s
171/s best
EloLMArena rating and rank.
—
1403
ReleasedMonth of release.
2026-11
2026-11
Input / 1MUS dollars per 1M input tokens.
$0.10 best
$0.30
Output / 1MUS dollars per 1M output tokens.
$0.40 best
$1.00
1K chats2K tokens in, 500 out, each.
$0.40 best
$1.10
1K documents20K tokens in, 1K out, each.
$2.40 best
$7.00
1K agent tasks200K tokens in, 20K out, each.
$28.00 best
$80.00
ContextMaximum tokens in one request.
1.05M
1.05M
AgenticMulti-step tasks with tools.
60.5
66.3 best
CodingCode writing and repair.
60.0
65.7 best
ReasoningLogic problems and puzzles.
61.6
67.0 best
MultimodalTasks with images and text.
57.1
63.0 best
KnowledgeFacts and expert knowledge.
49.6
61.1 best
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
—
—
MathMath problems.
47.4
65.0 best

Benchmarks

29 shared
Only Inkling 2 Mini: 7Only Inkling 2 Small: 33
Inkling 2 Mini
Inkling 2 Small
20.5%129 / 170
33.8% best73 / 170
86.6%47 / 144
93.0% best13 / 144
49.3%55 / 137
81.7% best23 / 137
84.5%56 / 132
96.5% best14 / 132
51.2%48 / 127
84.2% best22 / 127
90.5%49 / 116
97.4% best22 / 116
79.2%24 / 114
82.2% best11 / 114
71.6%51 / 89
85.1% best19 / 89
86.6%53 / 87
93.0% best14 / 87
DeepSWEAgentic
48.2%56 / 64
72.1% best19 / 64
BrowseCompAgentic
79.0%31 / 63
90.2% best9 / 63
74.9%39 / 58
81.6% best7 / 58
30.7%41 / 55
44.1% best18 / 55
MCP AtlasAgentic
82.7%12 / 53
85.0% best5 / 53
88.3% best15 / 50
87.7%17 / 50
JobBenchAgentic
49.6%17 / 37
70.5% best2 / 37
15.1%31 / 32
22.6% best24 / 32
46.3%17 / 26
51.2% best10 / 26
85.3%12 / 25
90.9% best9 / 25
LongBench v2Reasoning
66.9%5 / 25
70.6% best1 / 25