We propose SmoothQuant, a training-free, accuracy-preserving, and general-purpose post-training quantization (PTQ) solution to enable 8-bit weight, 8-bit activation (W8A8) quantization for LLMs.
EIE proposed to accelerate pruned and compressed neural networks, exploiting weight sparsity, activation sparsity, and 4-bit weight-sharing in neural network accelerators.
Low-bit weight-only quantization for LLMs.
In MCUNetV3, we enable on-device training under 256KB memory, using less than 1/1000 memory of PyTorch while matching the accuracy on the visual wake words application using system-algorithm co-design.
In MCUNetV2, we propose a generic patch-by-patch inference scheduling, which operates only on a small spatial region of the feature map and significantly cuts down the peak memory. We further propose network redistribution to shift the receptive field and FLOPs to the later stage and reduce the computation overhead.
MCUNet is a system-algorithm co-design framework for tiny deep learning on microcontrollers. It consists of TinyNAS and TinyEngine. They are co-designed to fit the tight memory budgets. With system-algorithm co-design, we can significantly improve the deep learning performance on the same tiny memory budget.
Running large language models (LLMs) on the edge is of great importance. In this blog, we introduce TinyChat, an efficient and lightweight system for LLM deployment on the edge. It runs Meta's latest LLaMA-2 model at 30 tokens / second on NVIDIA Jetson Orin and can easily support different models and hardware.
In MCUNetV3, we enable on-device training under 256KB SRAM and 1MB Flash, using less than 1/1000 memory of PyTorch while matching the accuracy on the visual wake words application. It enables the model to adapt to newly collected sensor data and users can enjoy customized services without uploading the data to the cloud thus protecting privacy.