Equationliu / KangarooLinks
[NeurIPS 2024] The official implementation of "Kangaroo: Lossless Self-Speculative Decoding for Accelerating LLMs via Double Early Exiting"
☆57Updated 11 months ago
Alternatives and similar repositories for Kangaroo
Users that are interested in Kangaroo are comparing it to the libraries listed below
Sorting:
- GEAR: An Efficient KV Cache Compression Recipefor Near-Lossless Generative Inference of LLM☆163Updated 10 months ago
- [ICLR 2025] PEARL: Parallel Speculative Decoding with Adaptive Draft Length☆83Updated last month
- QAQ: Quality Adaptive Quantization for LLM KV Cache☆50Updated last year
- Multi-Candidate Speculative Decoding☆35Updated last year
- Code associated with the paper **Draft & Verify: Lossless Large Language Model Acceleration via Self-Speculative Decoding**☆186Updated 3 months ago
- The Official Implementation of Ada-KV: Optimizing KV Cache Eviction by Adaptive Budget Allocation for Efficient LLM Inference☆76Updated 4 months ago
- Ouroboros: Speculative Decoding with Large Model Enhanced Drafting (EMNLP 2024 main)☆106Updated 2 months ago
- ☆248Updated last year
- ☆54Updated last year
- ☆50Updated 6 months ago
- ☆76Updated last month
- Official Pytorch Implementation of "Outlier Weighed Layerwise Sparsity (OWL): A Missing Secret Sauce for Pruning LLMs to High Sparsity"☆66Updated 11 months ago
- An innovative method expediting LLMs via streamlined semi-autoregressive generation and draft verification.☆26Updated last month
- 16-fold memory access reduction with nearly no loss☆94Updated 2 months ago
- ☆84Updated 3 years ago
- ☆129Updated 3 months ago
- PyTorch implementation of paper "Response Length Perception and Sequence Scheduling: An LLM-Empowered LLM Inference Pipeline".☆86Updated 2 years ago
- ☆42Updated 6 months ago