Skip to content
Leaderboard / Thinking Machines

Inkling

Thinking Machines OPEN Apache 2.0
Verdict

Inkling is a solid mid-pack model: #142 of 406 by SIscope score. Strongest at math, weaker at agents & tools. Mid-range on price. Weights are open, so you can run it yourself.

SIscope score
59
#142 of 406
Human-vote rank
#72
of 419 · score 1443
Price per 1M tokens, blended
$1.76
$1 in · $4.05 out
Context window
524K
524,288 tokens

Shape

Each spoke is a category, 0 at the centre, 100 at the edge.

CodingAgentsReasoningMathChatInstructions

Category scores

Percentiles: 90 means better than about 90% of tested models.

Coding60
Agents & tools4
Reasoning & knowledge83
Math92
Chat quality83
Following instructions81

What goes into each category · Compare with Gemini 4 Carbon

Human-vote score over time

Text arena, recent updates.

143514431450Oct 20: 1441.8 (rank #72)Oct 22: 1442 (rank #72)Oct 28: 1442.7 (rank #72)Oct 20Oct 28

Coding votes over time

WebDev arena, recent updates.

140514131420Oct 16: 1413.7 (rank #87)Oct 18: 1413.5 (rank #88)Oct 19: 1413.9 (rank #90)Oct 22: 1415 (rank #91)Oct 23: 1414.9 (rank #92)Oct 27: 1414.9 (rank #93)Oct 28: 1415.2 (rank #94)Oct 16Oct 28

Independent test runs

Run by Epoch AI on the same harness for every model. Bars show the share of questions answered correctly.

No Epoch AI runs for this model yet.

Human-vote arenas

People compare two anonymous answers and pick the better one. Scores work like chess ratings.

ArenaRankScoreRange (95%)VotesUpdated
Text arena
Everyday chat, blind human votes
#72 of 41914431439–144745,472Oct 28, 2026
WebDev arena
Building web apps, blind human votes
#94 of 15214151409–142117,073Oct 28, 2026
Vision arena
Questions about images, blind human votes
Not on this board yet
Agent arena
Real multi-step agent sessions
#55 of 57-16.1%-17.6–-14.6%44,251Nov 2, 2026

Source: Arena (formerly LMArena) leaderboard dataset, CC BY 4.0. Fetched Nov 4, 2026.

Price

Input
$1
per 1M tokens
Output
$4.05
per 1M tokens
Blended
$1.76
3 in : 1 out, per 1M
Context window
524K
max tokens per request
Speed
Coming soon
our own tests

List price from the OpenRouter models API (thinkingmachines/inkling), Nov 4, 2026. Providers can charge differently.