optimum and optimum-graphcore
Optimum-graphcore is a specialized hardware backend plugin for Optimum that enables transformer training specifically on Graphcore IPUs, making them complementary tools designed to be used together rather than alternatives.
About optimum
huggingface/optimum
🚀 Accelerate inference and training of 🤗 Transformers, Diffusers, TIMM and Sentence Transformers with easy to use hardware optimization tools
Supports hardware-specific backends including ONNX Runtime, OpenVINO, TensorRT-LLM, AWS Neuron, and Intel Gaudi through modular installations, enabling optimized inference across diverse accelerators. Provides unified APIs for model export, quantization, and graph optimization while maintaining compatibility with PyTorch, enabling deployment from research to production without refactoring model code.
About optimum-graphcore
huggingface/optimum-graphcore
Blazing fast training of 🤗 Transformers on Graphcore IPUs
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work