FB (Facebook) + GEMM (General Matrix-Matrix Multiplication) - https://code.fb.com/ml-applications/fbgemm/

Tensors and Dynamic neural networks in Python with strong GPU acceleration

🤗 Transformers: State-of-the-art Machine Learning for Pytorch, TensorFlow, and JAX.

Seamless operability between C++11 and Python

Open standard for machine learning interoperability

NVIDIA Federated Learning Application Runtime Environment

Label Studio is a multi-type data labeling and annotation tool with standardized output format

💥 Fast State-of-the-Art Tokenizers optimized for Research and Production

Collective communications library with various primitives for multi-machine training.

A tensor-aware point-to-point communication primitive for machine learning

nanobind: tiny and efficient C++/Python bindings

FlatBuffers: Memory Efficient Serialization Library

A CPU+GPU Profiling library that provides access to timeline traces and hardware performance counters.

CPU INFOrmation library (x86/x86-64/ARM/ARM64, Linux/Windows/Android/macOS/iOS)

Prototype for ONNX spec

Protocol Buffers - Google's data interchange format

A high-throughput and memory-efficient inference and serving engine for LLMs