Tags: #gpu

A high-performance, scalable JAX-based open-source library for training large language models on Google Cloud TPUs and GPUs.

An ultrafast, open-source Pythonic runtime for deploying and scaling serverless GPU inference, sandboxes, and background jobs with zero infrastructure overhead.

A Dockerized FastAPI wrapper providing a high-performance, multi-platform (CPU/GPU) and multi-language API for the Kokoro-82M text-to-speech model, compatible with OpenAI's speech endpoint.

PostgresML integrates machine learning and AI capabilities, including GPU acceleration and large language models, directly into PostgreSQL, eliminating the need for separate systems and data transfers.

A comprehensive open-source application for Linux that provides detailed monitoring and advanced configuration, including overclocking, for AMD, Nvidia, and Intel GPUs.