Release2026-07-15

Thinking Machines Lab released Inkling, its first open-weights model, on July 15, 2026. Inkling is a Mixture-of-Experts transformer with 975B total and 41B active parameters, a context window up to 1M tokens, pretrained on 45 trillion tokens of text, images, audio and video, with an encoder-free multimodal design (dMel spectrograms for audio, 40x40 pixel patches with a four-layer hMLP for vision). It offers a controllable thinking-effort setting from 0.2 to 0.99, and the company reports matching Nemotron 3 Ultra on Terminal Bench 2.1 at roughly a third of the tokens, plus scores of 91.4% on VoiceBench and 73.5% on MMMU Pro. Weights are on Hugging Face and the model is available to fine-tune on Tinker, with a new Inkling Playground in the Tinker console. A preview of Inkling-Small, with 12B active parameters, was also shared. The company states Inkling is not the strongest model available, open or closed.

Send this to someone who needs it

Shares the story and its sources. Nothing about you.

What does this mean for your job?

This is the story as everyone gets it. Once a week we send you the version written for your role — what changed, why it matters for the work you actually do, and one thing to try. Free while we tune it.