ranggihwang / Pregated_MoELinks
☆49Updated last year
Alternatives and similar repositories for Pregated_MoE
Users that are interested in Pregated_MoE are comparing it to the libraries listed below
Sorting:
- ArkVale: Efficient Generative LLM Inference with Recallable Key-Value Eviction (NIPS'24)☆42Updated 7 months ago
- LLM Inference analyzer for different hardware platforms☆82Updated 3 weeks ago
- ☆150Updated last year
- ☆73Updated 2 months ago
- MAGIS: Memory Optimization via Coordinated Graph Transformation and Scheduling for DNN (ASPLOS'24)☆53Updated last year
- A GPU-optimized system for efficient long-context LLMs decoding with low-bit KV cache.☆56Updated last week
- Explore Inter-layer Expert Affinity in MoE Model Inference☆12Updated last year
- ☆172Updated last year
- SpInfer: Leveraging Low-Level Sparsity for Efficient Large Language Model Inference on GPUs☆50Updated 4 months ago
- ☆23Updated last year
- ☆64Updated last year
- [HPCA'24] Smart-Infinity: Fast Large Language Model Training using Near-Storage Processing on a Real System☆46Updated 2 weeks ago
- Tender: Accelerating Large Language Models via Tensor Decompostion and Runtime Requantization (ISCA'24)☆20Updated last year
- InfiniGen: Efficient Generative Inference of Large Language Models with Dynamic KV Cache Management (OSDI'24)☆147Updated last year
- LLMServingSim: A HW/SW Co-Simulation Infrastructure for LLM Inference Serving at Scale☆125Updated 2 weeks ago
- ☆21Updated 4 months ago
- ☆54Updated last year
- [ICML 2024 Oral] Any-Precision LLM: Low-Cost Deployment of Multiple, Different-Sized LLMs☆111Updated last month
- ☆48Updated last month
- ☆67Updated last year
- ☆14Updated last year
- ☆51Updated last year
- A lightweight design for computation-communication overlap.☆155Updated last month
- Flash-LLM: Enabling Cost-Effective and Highly-Efficient Large Generative Model Inference with Unstructured Sparsity☆216Updated last year
- FlashSparse significantly reduces the computation redundancy for unstructured sparsity (for SpMM and SDDMM) on Tensor Cores through a Swa…☆29Updated 3 weeks ago
- Code release for AdapMoE accepted by ICCAD 2024☆29Updated 3 months ago
- A Vectorized N:M Format for Unleashing the Power of Sparse Tensor Cores☆52Updated last year
- ☆24Updated 8 months ago
- Magicube is a high-performance library for quantized sparse matrix operations (SpMM and SDDMM) of deep learning on Tensor Cores.☆89Updated 2 years ago
- ☆20Updated 4 months ago