Thinking Machines Lab Releases Inkling-Small: A 276B Total, 12B Active Open Weights Multimodal MoE Model
8/10Thinking Machines Lab released Inkling-Small, a multimodal Mixture-of-Experts (MoE) AI model featuring 276 billion total weights and 12 billion active open weights. Remarkably, this model can operate on a single NVIDIA B300 GPU, underscoring advances in efficiency and deployment of large-scale AI models.
