Inkling-Small vs Step 3.7 Flash

Change models
Inkling-SmallThinking Machines Lab
IndexAll categories, out of 100.
55.1 ±3.2
55.3 ±5.0 best
GapIndex points below the top.
−0.2 · no clear gap
Top
CoverageIndex weight that has results.
95% best
85%
SpeedOutput tokens per second.
80/s
112/s best
EloLMArena rating and rank.
1414
—
ReleasedMonth of release.
2026-07
2026-05
Input / 1MUS dollars per 1M input tokens.
$0.45
$0.20 best
Output / 1MUS dollars per 1M output tokens.
$1.20
$1.15 best
1K chats2K tokens in, 500 out, each.
$1.50
$0.975 best
1K documents20K tokens in, 1K out, each.
$10.20
$5.15 best
1K agent tasks200K tokens in, 20K out, each.
$114.00
$63.00 best
ContextMaximum tokens in one request.
524K best
262K
AgenticMulti-step tasks with tools.
55.7
59.6 best
CodingCode writing and repair.
54.9
54.9
ReasoningLogic problems and puzzles.
55.7 best
49.6
MultimodalTasks with images and text.
54.7
59.3 best
KnowledgeFacts and expert knowledge.
52.0 best
46.8
MultilingualTasks in many languages.
—
—
InstructionTasks with strict prompt rules.
65.8 best
51.9
MathMath problems.
55.3
—

Benchmarks

11 shared
Only Inkling-Small: 53Only Step 3.7 Flash: 10
Inkling-Small
Step 3.7 Flash
33.3% best70 / 222
21.4%110 / 222
75.7% best92 / 217
73.7%100 / 217
8.3% best74 / 214
2.3%104 / 214
89.5% best43 / 199
80.9%97 / 199
31.8% best74 / 164
25.8%101 / 164
31.2% best60 / 133
25.8%69 / 133
49.7% best53 / 114
43.9%78 / 114
74.0%61 / 113
75.3% best52 / 113
64.7% best59 / 86
59.5%65 / 86
55.9%52 / 83
56.3% best49 / 83
BrowseCompAgentic
77.4% best33 / 63
75.8%37 / 63