CentML

CentML is a technology company that provides a range of products designed to optimize the performance of AI models and reduce operational costs. Their offerings include CServe, Hidet, and DeepView. CServe, tailored for the LLM era, promises up to 70% cost reduction and up to 8x latency improvement in model serving. Hidet, an open-source deep learning compiler, enables end-to-end compilation of DNN models from PyTorch and ONNX to efficient CUDA kernels, facilitating enhanced performance. DeepView, another open-source tool, assists machine learning practitioners in identifying bottlenecks in models and predicting deployment time and costs. CentML claims their solutions can accelerate model inference by up to 8x and significantly cut down AI operation costs. The company has offices located in Toronto, Canada, and Palo Alto, California, positioning it strategically in North America’s tech hubs.