CAS-ICT HPDPS Lab

Pinned Loading

1. KVServe KVServe Public

   Service-aware KV-cache compression for bandwidth-efficient disaggregated LLM serving.

   Python 32 9

2. ElasticMM ElasticMM Public

   ElasticMM: Elastic and Efficient MLLM Serving System

   Python 47 2

3. COCCL COCCL Public

   Forked from NVIDIA/nccl

   COCCL: Compression and precision co-aware collective communication library

   C++ 40 5

4. PRISM PRISM Public

   An Efficient GPU-Based Lossy Compression Framework for Progressive Data Retrieval

   Cuda 5 4

5. ENEC ENEC Public

   A lossless encoder for AI models on Ascend NPUs.

   C++ 5 2

6. PhotonZip PhotonZip Public

   PhotonZip is a compression software package for light source data

   Python 4