Pinned Loading
Repositories
Showing 10 of 21 repositories
- BitBLAS-Quant Public Forked from wzqvip/BitBLAS-blackwell
BitBLAS is a library to support mixed-precision matrix multiplications, especially for quantized LLM deployment.
AIoT-MLSys-Lab/BitBLAS-Quant’s past year of commit activity - MEDA Public
[NAACL 2025🔥] MEDA: Dynamic KV Cache Allocation for Efficient Multimodal Long-Context Inference
AIoT-MLSys-Lab/MEDA’s past year of commit activity - Famba-V Public
[ECCV 2024 Workshop Best Paper Award] Famba-V: Fast Vision Mamba with Cross-Layer Token Fusion
AIoT-MLSys-Lab/Famba-V’s past year of commit activity