Open Source · MarkTechPost ·
Thinking Machines Lab Releases Inkling-Small: A 276B Total, 12B Active Open Weights Multimodal MoE Model
Thinking Machines Lab released Inkling-Small, an open-weights multimodal mixture-of-experts model with 276B total parameters and 12B active parameters. The company says it matches Inkling at one-quarter the size, while its NVFP4 checkpoint runs on a single NVIDIA B300 GPU.