> Inkling is an open-weights AI model from Thinking Machines, released 15 Jul 2026. Index score 150 (GPT-4 = 100), rank 59 of 229 ranked models. API price $1.0 in / $4.05 out per million tokens, 524K context.

[Models](https://themodelindex.org/models/) /[Thinking Machines](https://themodelindex.org/labs/thinking-machines/)

TM

# Inkling

Thinking Machines

Released 15 Jul 2026

Open

Date verified

[Compare](https://themodelindex.org/compare/inkling-vs-claude-opus-5/) [Run it yourself →](https://themodelindex.org/local/) [Announcement](https://thinkingmachines.ai/news/introducing-inkling/)

150139–161

Index score

#59 of 229 models

4/ 8

Core tests taken

4 of 6 areas · plus 12 anchor tests

−35

Versus the frontier at release

best then: Claude Fable 5

$1.0/ $4.05

Price per million tokens

input / output, list price

524K

Context window

tokens it reads at once

## Where it sits · on the frontier

_Chart: Model Index score of every scored model since late 2022, with the best closed and best open model over time_

GPT-5 reached this score 11 months earlier.

## Test results · by area · tick = best by any model

Show 12 anchor tests

Reasoning & knowledge

evidence

GPQA Diamond

88%

Humanity's Last Exam

no result

Maths

evidence

FrontierMath T1-3

33%

Coding

evidence

SWE-bench Verified

no result

Terminal-Bench

no result

Agents

evidence

APEX-Agents

34%

Novel problems

evidence

ARC-AGI-2

37%

Long tasks

no evidence yet

METR time horizon

no result

Independent results from the Epoch AI benchmarking hub, best reasoning setting. The core tests are shown here; anchor tests also inform the score and can be shown above.

## What others say · rank on each leaderboard

The Model Index

independent tests, our method

#59

of 219 · 150

LMArena · Maths

maths questions in chat

#34

of 373 · 1479

Artificial Analysis Intelligence Index

overall capability across ten evaluations

#43

of 207 · 25

LMArena · Text

overall chat quality, judged by people

#82

of 388 · 1442

LMArena · Coding

coding questions in chat

#82

of 383 · 1492

Epoch Capabilities Index

overall capability from many benchmarks

#57

of 239 · 148.6

LMArena · Creative writing

creative writing

#125

of 386 · 1379

LMArena · WebDev

building web apps and front ends

#72

of 123 · 1412

Further left is better. Leaderboards measure different things: LMArena is people voting blind between two answers, the others are test scores. Click any row for the source.

## Best at · top three among current models

Not in the top three among current models on any category we track.

Where this model places in the top three of today's models, on category leaderboards and on our core tests.

## Where to use it · prices checked 1 Oct 2026

#### From the lab

TM

Tinker

API

↗

#### API providers · 2 · per million tokens, in / out

D

DeepInfra

$0.95 / $4.05 · fp8

cheapest

T

Together

$1.0 / $4.05

↗

#### Run it yourself

🤗

Official weights

thinkingmachines/Inkling

↗

Provider prices come from OpenRouter's live listing; going direct to a provider can differ. Subscriptions are the lab's own apps.

## Facts and sources

Released

15 Jul 2026

Announcement

[thinkingmachines.ai ↗](https://thinkingmachines.ai/news/introducing-inkling/)

Release entry

Inkling

Scored as

Inkling

Artificial Analysis index

25 · xhigh

Weights

[huggingface.co/thinkingmachines/Inkling ↗](https://huggingface.co/thinkingmachines/Inkling)

Output speed

169 tokens/s

Hugging Face

[thinkingmachines/Inkling ↗](https://huggingface.co/thinkingmachines/Inkling)

## Thinking Machines releases · around this one

Inkling

Open

15 JUL 2026

index 150

Inkling-Small

Open

30 JUL 2026

index 154

---
Source: https://themodelindex.org/models/inkling/ · The Model Index · data as of 2026-10-01