Equationliu / Kangaroo
[NeurIPS 2024] The official implementation of "Kangaroo: Lossless Self-Speculative Decoding for Accelerating LLMs via Double Early Exiting"
☆51Updated 8 months ago
Alternatives and similar repositories for Kangaroo:
Users that are interested in Kangaroo are comparing it to the libraries listed below
- [ICLR 2025] PEARL: parallel speculative decoding with adaptive draft length☆52Updated this week
- Multi-Candidate Speculative Decoding☆34Updated 10 months ago
- QAQ: Quality Adaptive Quantization for LLM KV Cache☆47Updated 11 months ago
- GEAR: An Efficient KV Cache Compression Recipefor Near-Lossless Generative Inference of LLM☆157Updated 8 months ago
- 16-fold memory access reduction with nearly no loss☆80Updated 3 weeks ago
- The Official Implementation of Ada-KV: Optimizing KV Cache Eviction by Adaptive Budget Allocation for Efficient LLM Inference☆67Updated last month
- ☆44Updated 3 months ago
- ☆52Updated 11 months ago
- ☆229Updated 10 months ago
- Ouroboros: Speculative Decoding with Large Model Enhanced Drafting (EMNLP 2024 main)☆85Updated 5 months ago
- ☆38Updated 3 months ago
- Code associated with the paper **Draft & Verify: Lossless Large Language Model Acceleration via Self-Speculative Decoding**☆171Updated last month
- This repo contains the source code for: Model Tells You What to Discard: Adaptive KV Cache Compression for LLMs☆34Updated 6 months ago
- Spec-Bench: A Comprehensive Benchmark and Unified Evaluation Platform for Speculative Decoding (ACL 2024 Findings)☆238Updated this week
- The official implementation of paper: SimLayerKV: A Simple Framework for Layer-Level KV Cache Reduction.☆43Updated 4 months ago
- ☆49Updated 10 months ago
- Boosting 4-bit inference kernels with 2:4 Sparsity☆67Updated 6 months ago
- PyTorch implementation of paper "Response Length Perception and Sequence Scheduling: An LLM-Empowered LLM Inference Pipeline".☆82Updated last year
- ☆42Updated last year
- ☆115Updated 3 weeks ago
- Unofficial implementations of block/layer-wise pruning methods for LLMs.☆64Updated 10 months ago
- Official PyTorch implementation of IntactKV: Improving Large Language Model Quantization by Keeping Pivot Tokens Intact☆42Updated 9 months ago
- [ICML 2024] KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cache☆279Updated last month
- Fast and Robust Early-Exiting Framework for Autoregressive Language Models with Synchronized Parallel Decoding (EMNLP 2023 Long)☆56Updated 5 months ago
- [ICML 2024] Quest: Query-Aware Sparsity for Efficient Long-Context LLM Inference☆253Updated 3 months ago
- Activation-aware Singular Value Decomposition for Compressing Large Language Models☆57Updated 4 months ago
- [ICLR 2024] Jaiswal, A., Gan, Z., Du, X., Zhang, B., Wang, Z., & Yang, Y. Compressing llms: The truth is rarely pure and never simple.☆21Updated last year
- Sirius, an efficient correction mechanism, which significantly boosts Contextual Sparsity models on reasoning tasks while maintaining its…☆21Updated 6 months ago
- [ICLR2025] Breaking Throughput-Latency Trade-off for Long Sequences with Speculative Decoding☆109Updated 3 months ago