hpcaitech / CachedEmbedding
A memory efficient DLRM training solution using ColossalAI
☆95Updated last year
Related projects: ⓘ
- ☆108Updated 6 months ago
- A Python library transfers PyTorch tensors between CPU and NVMe☆92Updated last year
- Inference framework for MoE layers based on TensorRT with Python binding☆41Updated 3 years ago
- ☆83Updated 3 weeks ago
- ☆50Updated 3 months ago
- ☆105Updated last week
- Modular and structured prompt caching for low-latency LLM inference☆43Updated 4 months ago
- ☆70Updated 9 months ago
- A MoE impl for PyTorch, [ATC'23] SmartMoE☆56Updated last year
- ☆32Updated last week
- Official repository for LightSeq: Sequence Level Parallelism for Distributed Training of Long Context Transformers☆183Updated last month
- Compare different hardware platforms via the Roofline Model for LLM inference tasks.☆71Updated 6 months ago
- USP: Unified (a.k.a. Hybrid, 2D) Sequence Parallel Attention for Long Context Transformers Model Training and Inference☆309Updated this week
- PyTorch library for cost-effective, fast and easy serving of MoE models.☆90Updated last month
- ATC23 AE☆42Updated last year
- Odysseus: Playground of LLM Sequence Parallelism☆50Updated 3 months ago
- A collection of models built with ColossalAI☆32Updated last year
- Summary of system papers/frameworks/codes/tools on training or serving large model☆56Updated 9 months ago
- An efficient GPU support for LLM inference with x-bit quantization (e.g. FP6,FP5).☆173Updated 3 months ago
- ☆60Updated last month
- A low-latency & high-throughput serving engine for LLMs☆174Updated last week
- Boosting 4-bit inference kernels with 2:4 Sparsity☆47Updated 2 weeks ago
- ☆205Updated last year
- GPTQ inference Triton kernel☆273Updated last year
- [MLSys'24] Atom: Low-bit Quantization for Efficient and Accurate LLM Serving☆258Updated 2 months ago
- Elixir: Train a Large Language Model on a Small GPU Cluster☆12Updated last year
- Low-bit optimizers for PyTorch☆109Updated 11 months ago
- A high-performance distributed deep learning system targeting large-scale and automated distributed training. If you have any interests, …☆101Updated 9 months ago
- [NeurIPS'23] H2O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models.☆360Updated last month
- PyTorch bindings for CUTLASS grouped GEMM.☆57Updated 2 months ago