candlehuggingface/candle
Candle is a minimalist ML framework for Rust with a focus on performance (including GPU support) and ease of use.
17 libraries, 9 updated this year
Rust's machine learning projects focus on fast, dependency-light inference. Hugging Face's candle is a minimal tensor framework for running models on CPU, CUDA and Metal; burn is a full framework that also trains, with swappable backends. Hugging Face's tokenizers library is written in Rust and powers its Python package.
17 libraries
Candle is a minimalist ML framework for Rust with a focus on performance (including GPU support) and ease of use.
Burn is a new comprehensive dynamic Deep Learning Framework built using Rust with extreme flexibility, compute efficiency and portability as its primary goals
Fast State-of-the-Art Tokenizers optimized for Research and Production
Python-free Rust inference server for NLP models with OpenAI API compatibility and hot model swapping.
Open source framework for machine intelligence, sharing concepts from TensorFlow and Caffe. Available under the MIT license.
Rust native ready-to-use NLP pipelines and transformer-based models (BERT, DistilBERT, GPT2,...)
Deep learning in Rust, with shape checked tensors and neural networks
A machine learning framework featuring logistic regression, support vector machines, decision trees and random forests.
Deeplearn-rs provides simple networks that use matrix multiplication, addition, and ReLU under the MIT license.
Pure-Rust tokenizer for GGUF models, compatible with llama.cpp tokenization.