Popular repositories Loading

1. DistServe DistServe Public

   Disaggregated serving system for Large Language Models (LLMs).

   Jupyter Notebook 809 94

2. SwiftTransformer SwiftTransformer Public

   High performance Transformer implementation in C++.

   C++ 154 18

3. dLoRA-artifact dLoRA-artifact Public

   Jupyter Notebook 32 8

4. FastServe FastServe Public

   Jupyter Notebook 27 4

5. hydraserve hydraserve Public

   Python 18 4

6. WarmServe WarmServe Public

   WarmServe is a multi-LLM serving system that prewarms models according to workload predictions.

   Python 2