Infini-AI-Lab / SiriusLinks
Sirius, an efficient correction mechanism, which significantly boosts Contextual Sparsity models on reasoning tasks while maintaining its efficiency gain.
☆22Updated 9 months ago
Alternatives and similar repositories for Sirius
Users that are interested in Sirius are comparing it to the libraries listed below
Sorting:
- ☆26Updated 7 months ago
- [ICLR 2025] TidalDecode: A Fast and Accurate LLM Decoding with Position Persistent Sparse Attention☆39Updated 2 months ago
- The official implementation of paper: SimLayerKV: A Simple Framework for Layer-Level KV Cache Reduction.☆46Updated 8 months ago
- 16-fold memory access reduction with nearly no loss☆99Updated 2 months ago
- ☆37Updated 9 months ago
- SQUEEZED ATTENTION: Accelerating Long Prompt LLM Inference☆47Updated 7 months ago
- Pytorch implementation of our paper accepted by ICML 2024 -- CaM: Cache Merging for Memory-efficient LLMs Inference☆39Updated last year
- ☆30Updated last month
- ☆130Updated 4 months ago
- GEAR: An Efficient KV Cache Compression Recipefor Near-Lossless Generative Inference of LLM☆163Updated 11 months ago
- Activation-aware Singular Value Decomposition for Compressing Large Language Models☆71Updated 8 months ago
- ☆52Updated 6 months ago
- [ICLR2025] Breaking Throughput-Latency Trade-off for Long Sequences with Speculative Decoding☆116Updated 6 months ago
- Official Implementation of SAM-Decoding: Speculative Decoding via Suffix Automaton☆28Updated 4 months ago
- PoC for "SpecReason: Fast and Accurate Inference-Time Compute via Speculative Reasoning" [arXiv '25]☆39Updated last month
- This repo contains the source code for: Model Tells You What to Discard: Adaptive KV Cache Compression for LLMs☆38Updated 10 months ago
- Beyond KV Caching: Shared Attention for Efficient LLMs☆19Updated 11 months ago
- ☆104Updated 2 weeks ago
- [NeurIPS 2024] The official implementation of "Kangaroo: Lossless Self-Speculative Decoding for Accelerating LLMs via Double Early Exitin…☆57Updated 11 months ago
- [NeurIPS'23] Speculative Decoding with Big Little Decoder☆92Updated last year
- Boosting 4-bit inference kernels with 2:4 Sparsity☆79Updated 9 months ago
- Code for paper: [ICLR2025 Oral] FlexPrefill: A Context-Aware Sparse Attention Mechanism for Efficient Long-Sequence Inference☆113Updated last month
- ☆39Updated 7 months ago
- Code for "RSQ: Learning from Important Tokens Leads to Better Quantized LLMs"☆18Updated last week
- ☆50Updated last year
- [ICLR 2024 Spotlight] Code for the paper "Merge, Then Compress: Demystify Efficient SMoE with Hints from Its Routing Policy"☆85Updated last year
- The official implementation of the paper "Towards Efficient Mixture of Experts: A Holistic Study of Compression Techniques (TMLR)".☆71Updated 3 months ago
- Easy control for Key-Value Constrained Generative LLM Inference(https://arxiv.org/abs/2402.06262)☆63Updated last year
- ☆42Updated 7 months ago
- Ouroboros: Speculative Decoding with Large Model Enhanced Drafting (EMNLP 2024 main)☆106Updated 3 months ago