Intel / article

Thinking Machines introduces Inkling, an open-weight 975B MoE model

Why it matters

Inkling adds a large open-weight option for long-context, multimodal, and custom model workloads, changing the build-versus-API decision for professional teams.
Thinking Machines introduced Inkling, a multimodal mixture-of-experts model with 975B total parameters, 41B active parameters, up to a one-million-token context window, open weights, and customization support.