On November 17, Shangtang Technology officially released SenseNova Loop-9B. The total number of parameters for this model is approximately 9B. By executing the shared middle layer twice, the model expands the 32 physical decoding layers to 56 sub-layers of computation; supports text, image input, and text output, and the context length can reach up to 512K tokens.

Zhitongcaijing · 2d ago
On November 17, Shangtang Technology officially released SenseNova Loop-9B. The total number of parameters for this model is approximately 9B. By executing the shared middle layer twice, the model expands the 32 physical decoding layers to 56 sub-layers of computation; supports text, image input, and text output, and the context length can reach up to 512K tokens.