Follow us on:
Logic Fruit Technologies, having itself established in traditional use cases of complex FPGA designs, is making strides in latest AI/ML and in-line processing use cases for heterogeneous applications. We provide optimized solutions for high performance and low latency by capitalizing our expertise on not just FPGA but CPU and GPU as well. Here’s the overview of our expertise and the solutions we provide:
Deep expertise in NLP, NLU an NLG, classification models and other CNN based models.
Working with latest AI/ML frameworks and toolsets like TensorFlow, TensorFlow Lite, Pytorch, Caffe, Keras, Darknet, JupyterNB etc.
Working with SDK’s and tools like Xilinx Vitis, Pynq, AI Optimizer, DNNK etc.
Expertise in porting AI/ML model to other frameworks and pruning.
Achieving performance using hardware parallelism and multi-threading.
Expert in providing hardware accelerated AI/ML solutions using FPGA and GPU.
Model deployment via Docker containers, as Apps for Xilinx App Store and Cloud.
What Logic Fruit has for you!
On-edge deployment on platforms like Alveo U200 and Ultrascale96.
Cloud deployment on AWS.
Hardware and model based DPU customization.
Custom Algorithm or kernel development in RTL and HLS (C/C++ and OpenCL) for FPGA
Accelize DRM integration with RTL IP’s.
Accelerate models using GPU’s.
AI/ML model benchmarking.
Porting models to other frameworks (TensorFlow, Caffe, Darknet, Pytorch etc.)
Application and hardware specific performance tuning and optimization of AI/ML models.
Custom model pruning.
Algorithm development and optimization.
Synthetic dataset generation.
Tools and Infrastructure
We work with innovative R&D leaders, project managers, and CTOs who need a proven, flexible partner that can provide extensive FPGA-centric expertise, prototyping capabilities, or another high throughput FPGA/SoC embedded solutions delivering a high-quality embedded solution they can trust.
or just Call us on