#thinking-machines

Open-weight multimodal models just gained a serious new entrant. Thinking Machines released Inkling with full weights, 64K and 256K context options, and a direct fine-tuning path through Tinker.

HN readers focused less on leaderboard dominance and more on the package: Thinking Machines Lab is offering a multimodal MoE with controllable reasoning effort and Tinker-based fine-tuning as an open-weight base.

Thinking Machines Lab said it signed a multi-year strategic partnership with NVIDIA to deploy at least one gigawatt of next-generation Vera Rubin systems. The companies also plan to co-design training and serving systems and widen access to frontier AI and open models for enterprises, research institutions, and the scientific community.

NVIDIAAI says it is partnering with Thinking Machines to deploy at least 1 gigawatt of NVIDIA Vera Rubin systems for frontier AI training. Thinking Machines frames the partnership as infrastructure for both frontier model training and platforms for customizable AI.