alexzhang13 / flashattention2-custom-mask
Triton implementation of FlashAttention2 that adds Custom Masks.
☆78Updated 3 months ago
Related projects ⓘ
Alternatives and complementary repositories for flashattention2-custom-mask
- Triton-based implementation of Sparse Mixture of Experts.☆185Updated last month
- ☆189Updated 6 months ago
- ☆96Updated 2 months ago
- Odysseus: Playground of LLM Sequence Parallelism☆57Updated 5 months ago
- Official repository for LightSeq: Sequence Level Parallelism for Distributed Training of Long Context Transformers☆195Updated 3 months ago
- [ICML 2024] Quest: Query-Aware Sparsity for Efficient Long-Context LLM Inference☆203Updated 3 weeks ago
- Sirius, an efficient correction mechanism, which significantly boosts Contextual Sparsity models on reasoning tasks while maintaining its…☆19Updated 2 months ago
- ☆132Updated last year
- Repository of the paper "Accelerating Transformer Inference for Translation via Parallel Decoding"☆110Updated 8 months ago
- [NeurIPS'23] Speculative Decoding with Big Little Decoder☆86Updated 9 months ago
- ☆45Updated 6 months ago
- Code for Palu: Compressing KV-Cache with Low-Rank Projection☆57Updated last week
- GEAR: An Efficient KV Cache Compression Recipefor Near-Lossless Generative Inference of LLM☆149Updated 4 months ago
- Low-bit optimizers for PyTorch☆119Updated last year
- Explorations into some recent techniques surrounding speculative decoding☆212Updated last year
- [ICML 2024] KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cache☆243Updated last month
- Spec-Bench: A Comprehensive Benchmark and Unified Evaluation Platform for Speculative Decoding (ACL 2024 Findings)☆190Updated 3 weeks ago
- 16-fold memory access reduction with nearly no loss☆58Updated last week
- A sparse attention kernel supporting mix sparse patterns☆58Updated last month
- [MLSys'24] Atom: Low-bit Quantization for Efficient and Accurate LLM Serving☆278Updated 4 months ago
- Code associated with the paper **Draft & Verify: Lossless Large Language Model Acceleration via Self-Speculative Decoding**☆140Updated 6 months ago
- PyTorch bindings for CUTLASS grouped GEMM.☆68Updated 4 months ago
- ☆88Updated 2 months ago
- Code for "Everybody Prune Now: Structured Pruning of LLMs with only Forward Passes"☆28Updated 7 months ago
- Homepage for ProLong (Princeton long-context language models) and paper "How to Train Long-Context Language Models (Effectively)"☆119Updated this week
- Ouroboros: Speculative Decoding with Large Model Enhanced Drafting (EMNLP 2024 main)☆77Updated last month
- The source code of our work "Prepacking: A Simple Method for Fast Prefilling and Increased Throughput in Large Language Models"☆56Updated last month
- This repo contains the source code for: Model Tells You What to Discard: Adaptive KV Cache Compression for LLMs☆32Updated 3 months ago
- Breaking Throughput-Latency Trade-off for Long Sequences with Speculative Decoding☆79Updated this week
- The official implementation of paper: SimLayerKV: A Simple Framework for Layer-Level KV Cache Reduction.☆39Updated last month