Skip to content

FS-MoE

Sparse mixture-of-experts inference IP

Built for dynamically activated sparse models, carrying large parameter counts within an edge bandwidth budget.

FS-MoE Sparse mixture-of-experts inference IP, illustration
Fig. 1 · FS-MoE Sparse mixture-of-experts inference IP

Data flow

Built for dynamically activated sparse models, carrying large parameter counts within an edge bandwidth budget.

RoutergatingExpert 1Expert 2Expert 3Expert 4CombineOnly selectedweights are read
Fig. 2 · FS-MoE data flow (schematic)

Native support

Large sparse models on edge budgets

Deployment

ASIC tape-out and FPGA prototyping, with interface and system adaptation per project.

Licence includes

Inference core IP, FPGA evaluation board, model mapping and interface adaptation, design-in support.

IP coverage, interfaces, resource usage, performance targets and licensing terms are evaluated per target device and model.

Ask about FS-MoE

Tell us the target device, model and interfaces. We reply with IP coverage and an evaluation plan.

Ask about IP licensing