PyyanThe AI Index

Pyyan / Open-Weight Models / Inkling

T

Inkling

975B under Apache 2.0, shipped as a starting point rather than a finished product.

currentThinking Machines Lab ↗Verified 2026-09-06

975B

41B active

Specifications

Context1Msource · as of 2026-09-06
Input ($/Mtok)Weights onlysource · as of 2026-09-06
Output ($/Mtok)Weights onlysource · as of 2026-09-06
LicenceApache 2.0source · as of 2026-09-06
Max outputNot publishedsource · as of 2026-09-06
Parameters975B total, 41B activesource · as of 2026-09-06
Released15 Jul 2026source · as of 2026-09-06
Model IDNot publishedsource · as of 2026-09-06

Official links

Inkling model pageThinking Machines Lab official site

What matters

  • Pretrained on 45 trillion tokens of text, images, audio and video, with an encoder free architecture
  • Apache 2.0, weights on Hugging Face, same day fine tuning through the lab's Tinker platform
  • The lab earns from the compute customers use to adapt it, not from per token access

Best for fine tuning on your own data.

Compare with

All 5 side by side →

Inkling vs Qwen3.8-Flash-Next →Inkling vs DeepSeek V4.1 Flash →Inkling vs DeepSeek V4-Pro →Inkling vs Atria Dawn Preview →Inkling vs DeepSeek V4 Flash Vision Exp →

Also in Open-Weight Models

Qwen3.8-Flash-Next · AlibabaDeepSeek V4.1 Flash · DeepSeekDeepSeek V4-Pro · DeepSeekAtria Dawn Preview · Shanghai AI LaboratoryDeepSeek V4 Flash Vision Exp · DeepSeek