Summary
Inkling-Small is an open-weights multimodal Mixture-of-Experts model from Thinking Machines Lab. It is designed to match much of the larger Inkling model's capability while using a smaller footprint and controllable reasoning effort.
Inkling-Small is an open-weights multimodal Mixture-of-Experts model from Thinking Machines Lab. It is designed to match much of the larger Inkling model's capability while using a smaller footprint and controllable reasoning effort.
The model targets reasoning, agentic tasks, multimodality, epistemic calibration, and safety. Thinking Machines positions it as a compact model that can be fine-tuned through Tinker, making it practical for teams that want to adapt an open model to domain-specific workflows.
Inkling-Small is useful for developers building multimodal assistants, research agents, fine-tuned enterprise tools, and evaluation pipelines. Its open-weights release gives teams more control over inspection, customization, and deployment than a closed hosted-only model.
Get more likes & reach the top of search results by adding this button on your site!


Get top updates in AI to your inbox every weekend. It's free!