yuzhenmao / IceFormer
Implementation of IceFormer: Accelerated Inference with Long-Sequence Transformers on CPUs (ICLR 2024).
☆22Updated 8 months ago
Alternatives and similar repositories for IceFormer:
Users that are interested in IceFormer are comparing it to the libraries listed below
- ☆30Updated 8 months ago
- Odysseus: Playground of LLM Sequence Parallelism☆64Updated 8 months ago
- [ICLR 2024] This is the official PyTorch implementation of "QLLM: Accurate and Efficient Low-Bitwidth Quantization for Large Language Mod…☆24Updated 11 months ago
- [ICLR 2024] This is the official PyTorch implementation of "QLLM: Accurate and Efficient Low-Bitwidth Quantization for Large Language Mod…☆35Updated 11 months ago
- Linear Attention Sequence Parallelism (LASP)☆77Updated 8 months ago
- 32 times longer context window than vanilla Transformers and up to 4 times longer than memory efficient Transformers.☆45Updated last year
- Benchmark tests supporting the TiledCUDA library.☆15Updated 3 months ago
- [ACL 2024] RelayAttention for Efficient Large Language Model Serving with Long System Prompts☆38Updated 11 months ago
- Transformers components but in Triton☆31Updated 3 months ago
- Official Code For Dual Grained Quantization: Efficient Fine-Grained Quantization for LLM☆14Updated last year
- Decoding Attention is specially optimized for multi head attention (MHA) using CUDA core for the decoding stage of LLM inference.☆29Updated 3 months ago
- ACL 2023☆38Updated last year
- [ICML 2024] When Linear Attention Meets Autoregressive Decoding: Towards More Effective and Efficient Linearized Large Language Models☆28Updated 8 months ago
- Official implementation of the ICML 2024 paper RoSA (Robust Adaptation)☆38Updated last year
- IntLLaMA: A fast and light quantization solution for LLaMA☆18Updated last year
- Repository for Sparse Finetuning of LLMs via modified version of the MosaicML llmfoundry☆40Updated last year
- ☆22Updated last year
- Repository for CPU Kernel Generation for LLM Inference☆25Updated last year
- BESA is a differentiable weight pruning technique for large language models.☆14Updated 11 months ago
- Quantized Attention on GPU☆34Updated 2 months ago
- Framework to reduce autotune overhead to zero for well known deployments.☆61Updated 3 weeks ago
- ☆18Updated last week
- SQUEEZED ATTENTION: Accelerating Long Prompt LLM Inference☆40Updated 3 months ago
- GPTQ inference TVM kernel☆38Updated 9 months ago
- TensorRT LLM Benchmark Configuration☆13Updated 6 months ago
- An auxiliary project analysis of the characteristics of KV in DiT Attention.☆25Updated 2 months ago
- ☆68Updated this week
- Efficient Mixture of Experts for LLM Paper List☆35Updated 2 months ago
- TileFusion is a highly efficient kernel template library designed to elevate the level of abstraction in CUDA C for processing tiles.☆56Updated this week