Latest SOTA models supported on Qualcomm NPU.
AI & ML interests
On Device AI Deployment and Research
Recent Activity
Multimodal models running on Qualcomm NPU for Qualcomm IQ9 and RB3
Nexa AI infra to support Qwen3VL running on GPU/NPU/CPU
Tiny, multimodal on-device models developed by Nexa AI.
Latest SOTA models supported on Qualcomm NPU.
Latest SOTA models supported on Apple Neural Engine
Multimodal models running on Qualcomm NPU for Qualcomm IQ9 and RB3
Multimodal models running on Qualcomm NPU for Snapdragon8 Gen4
Nexa AI infra to support Qwen3VL running on GPU/NPU/CPU
NexaQuant compresses models with 100% accuracy recovery.
Tiny, multimodal on-device models developed by Nexa AI.