Infini-AI-Lab / Sirius
Sirius, an efficient correction mechanism, which significantly boosts Contextual Sparsity models on reasoning tasks while maintaining its efficiency gain.
☆19Updated 2 months ago
Related projects ⓘ
Alternatives and complementary repositories for Sirius
- GEAR: An Efficient KV Cache Compression Recipefor Near-Lossless Generative Inference of LLM☆147Updated 4 months ago
- 16-fold memory access reduction with nearly no loss☆59Updated last week
- [NeurIPS'23] Speculative Decoding with Big Little Decoder☆86Updated 9 months ago
- ☆45Updated 6 months ago
- Fast and Robust Early-Exiting Framework for Autoregressive Language Models with Synchronized Parallel Decoding (EMNLP 2023 Long)☆53Updated last month
- Official Pytorch Implementation of "Outlier Weighed Layerwise Sparsity (OWL): A Missing Secret Sauce for Pruning LLMs to High Sparsity"☆51Updated 4 months ago
- Ouroboros: Speculative Decoding with Large Model Enhanced Drafting (EMNLP 2024 main)☆76Updated last month
- MagicPIG: LSH Sampling for Efficient LLM Generation☆58Updated 3 weeks ago
- Code for Palu: Compressing KV-Cache with Low-Rank Projection☆57Updated this week
- ☆188Updated 6 months ago
- Boosting 4-bit inference kernels with 2:4 Sparsity☆51Updated 2 months ago
- This repo contains the source code for: Model Tells You What to Discard: Adaptive KV Cache Compression for LLMs☆32Updated 3 months ago
- Triton-based implementation of Sparse Mixture of Experts.☆185Updated last month
- ☆96Updated last month
- The official implementation of paper: SimLayerKV: A Simple Framework for Layer-Level KV Cache Reduction.☆38Updated last month
- Implementation of Kangaroo: Lossless Self-Speculative Decoding via Double Early Exiting☆44Updated 4 months ago
- Breaking Throughput-Latency Trade-off for Long Sequences with Speculative Decoding☆78Updated this week
- A sparse attention kernel supporting mix sparse patterns☆55Updated last month
- TidalDecode: A Fast and Accurate LLM Decoding with Position Persistent Sparse Attention☆21Updated last month
- ☆20Updated last week
- [ICML 2024] Quest: Query-Aware Sparsity for Efficient Long-Context LLM Inference☆202Updated 2 weeks ago
- ☆88Updated 2 months ago
- ☆31Updated 2 months ago
- Official repository for LightSeq: Sequence Level Parallelism for Distributed Training of Long Context Transformers☆195Updated 3 months ago
- 📰 Must-read papers on KV Cache Compression (constantly updating 🤗).☆136Updated this week
- Activation-aware Singular Value Decomposition for Compressing Large Language Models☆49Updated 3 weeks ago
- Code associated with the paper **Draft & Verify: Lossless Large Language Model Acceleration via Self-Speculative Decoding**☆138Updated 5 months ago
- Pytorch implementation of our paper accepted by ICML 2024 -- CaM: Cache Merging for Memory-efficient LLMs Inference☆26Updated 5 months ago
- ☆31Updated this week
- The official implementation of the paper "Demystifying the Compression of Mixture-of-Experts Through a Unified Framework".☆48Updated 3 weeks ago