RWKV (pronounced RwaKuv) is an RNN with great LLM performance, which can also be directly trained like a GPT transformer (parallelizable). We are at RWKV-7 "Goose". So it's combining the best of RNN and transformer - great performance, linear time, constant space (no kv-cache), fast training, infinite ctx_len, and free sentence embedding.
Repository Index
47,133 repositories indexed
1–5 of 5 · Most stars
Open Machine Learning Compiler Framework
20+ high-performance LLMs with recipes to pretrain, finetune and deploy at scale.
YOLOX is a high-performance anchor-free YOLO, exceeding yolov3~v5 with MegEngine, ONNX, TensorRT, ncnn, and OpenVINO supported. Documentation: https://yolox.readthedocs.io/
Pythonic tool for orchestrating machine-learning/high performance/quantum-computing workflows in heterogeneous compute environments.