Flex Logix Technologies offers the InferX family of Inference Accelerators for use in Edge AI and Vision applications. Based on dynamically reconfigurable TPU technology, InferX-based products offer high performance, low power and efficient execution of complex deep learning models in real time at the Edge. InferX offers INT8, INT16 and BF16 data types and supports the compilation and execution of trained models described in .ONNX and .TFLITE formats. InferX X1 products are available in PCIe, M.2 and chip form-factors.
Flex Logix
