Inkling - Thinking Machines Lab

Open weights, ready to tinker

An efficient open model for text, images, and audio, available to fine-tune.

Model at a glance

A well-rounded, generalist model

Spider chart comparing Inkling, Nemotron 3 Ultra, GLM 5.2, GPT 5.6 Sol, and Claude Fable 5 on ten evaluations scored from zero to one hundred. Inkling is shown with the heavier cobalt line. Evaluations without a reported model score are plotted at zero. Hover an evaluation to compare every model's score.

Model HLE (no tools) Reasoning AIME 2026 Reasoning SWE-bench Pro (Public) Agentic (coding) Terminal Bench 2.1 Agentic (coding) MCP Atlas Agentic (general) SimpleQA Verified Factuality AA Omniscience (acc) Factuality Audio MCAudio MMMU Pro Vision IFBench Chat
Inkling 40 60 80 100 60 70 80 90 70 75 85 65 80 75 70 80 85 70 75 80
Nemotron 3 Ultra 30 50 60 90 50 60 70 80 50 65 75 55 75 70 65 75 80 60 70 70
GLM 5.2 30 40 50 70 40 40 50 60 40 50 60 50 65 60 50 65 70 50 60 65
GPT 5.6 Sol 20 30 40 60 30 30 40 50 30 40 50 40 55 50 40 55 60 40 50 50
Claude Fable 5 10 20 30 50 20 20 30 40 20 30 40 30 45 40 30 45 50 30 40 40

Key Features

Ready to customize

Fine-tune Inkling for your domain on Tinker, our training platform.