Gemini 3 Flash vs Inkling-Small

Change models
Inkling-SmallThinking Machines Lab
IndexAll categories, out of 100.
55.0 ±4.0
55.1 ±3.2 best
GapIndex points below the top.
−0.1 · no clear gap
Top
CoverageIndex weight that has results.
95%
95%
SpeedOutput tokens per second.
198/s best
80/s
EloLMArena rating and rank.
1466 best
1414
ReleasedMonth of release.
2025-12
2026-07
Input / 1MUS dollars per 1M input tokens.
$0.50
$0.45 best
Output / 1MUS dollars per 1M output tokens.
$3.00
$1.20 best
1K chats2K tokens in, 500 out, each.
$2.50
$1.50 best
1K documents20K tokens in, 1K out, each.
$13.00
$10.20 best
1K agent tasks200K tokens in, 20K out, each.
$160.00
$114.00 best
ContextMaximum tokens in one request.
1M best
524K
AgenticMulti-step tasks with tools.
48.5
55.7 best
CodingCode writing and repair.
56.8 best
54.9
ReasoningLogic problems and puzzles.
50.8
55.7 best
MultimodalTasks with images and text.
67.3 best
54.7
KnowledgeFacts and expert knowledge.
60.7 best
52.0
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
40.5
65.8 best
MathMath problems.
57.3 best
55.3

Benchmarks

12 shared
Only Gemini 3 Flash: 19Only Inkling-Small: 52
Gemini 3 Flash
Inkling-Small
15.0%127 / 222
33.3% best70 / 222
55.3%147 / 217
75.7% best92 / 217
89.4% best47 / 217
88.5%53 / 217
1.4%113 / 214
8.3% best74 / 214
95.6% best40 / 200
90.0%62 / 200
81.2%95 / 199
89.5% best43 / 199
45.8% best34 / 164
31.8%74 / 164
78.6% best33 / 113
74.0%61 / 113
66.8% best13 / 91
19.1%79 / 91
51.2% best54 / 90
46.3%57 / 90
20.0% best45 / 80
6.0%74 / 80
17.1%52 / 72
17.1%52 / 72