FS-Attention
Hybrid attention inference IP
Native support for language models built on hybrid attention, including the gated full-attention layers of the Qwen3.5 family.

Data flow
Native support for language models built on hybrid attention, including the gated full-attention layers of the Qwen3.5 family.
Native support
Qwen3 / 3.5 (0.6B to 9B) and SmolLM2
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.
Further reading
FPGA Prototyping for LLM Accelerators: IP Licence to Tape-outRunning an LLM inference core on an FPGA before tape-out validates model mapping, interfaces and the system path. What an IP licence covers, step by step.
Qwen3.5 to Qwen3.8: Why Hybrid Attention Suits Edge SiliconQwen3.5, 3.6 and 3.8 interleave Gated DeltaNet linear attention with gated full attention at 3:1. Why that layout keeps memory in check on edge AI silicon.
How to Choose an Edge LLM Inference Chip: 7 QuestionsChoosing an edge LLM inference chip comes down to model fit, power budget and how the model updates, not peak TOPS. Seven questions to answer, plus a checklist.
Ask about FS-Attention
Tell us the target device, model and interfaces. We reply with IP coverage and an evaluation plan.
