feifeibear / LLMSpeculativeSampling
Fast inference from large lauguage models via speculative decoding
β569Updated 2 months ago
Related projects β
Alternatives and complementary repositories for LLMSpeculativeSampling
- π° Must-read papers and blogs on Speculative Decoding β‘οΈβ471Updated last week
- Official Implementation of EAGLE-1 (ICML'24) and EAGLE-2 (EMNLP'24)β826Updated this week
- [NeurIPS'23] H2O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models.β391Updated 3 months ago
- β289Updated 7 months ago
- Spec-Bench: A Comprehensive Benchmark and Unified Evaluation Platform for Speculative Decoding (ACL 2024 Findings)β188Updated 3 weeks ago
- USP: Unified (a.k.a. Hybrid, 2D) Sequence Parallel Attention for Long Context Transformers Model Training and Inferenceβ357Updated this week
- Ring attention implementation with flash attentionβ585Updated last week
- [ICLR 2024] Sheared LLaMA: Accelerating Language Model Pre-training via Structured Pruningβ558Updated 8 months ago
- [ICML 2024] Break the Sequential Dependency of LLM Inference Using Lookahead Decodingβ1,149Updated last month
- FlashInfer: Kernel Library for LLM Servingβ1,452Updated this week
- Analyze the inference of Large Language Models (LLMs). Analyze aspects like computation, storage, transmission, and hardware roofline modβ¦β311Updated 2 months ago
- Explorations into some recent techniques surrounding speculative decodingβ211Updated last year
- [ACL 2024] LongBench: A Bilingual, Multitask Benchmark for Long Context Understandingβ671Updated 2 months ago
- REST: Retrieval-Based Speculative Decoding, NAACL 2024β176Updated last month
- QServe: W4A8KV4 Quantization and System Co-design for Efficient LLM Servingβ443Updated last week
- β·οΈ LLaMA-MoE: Building Mixture-of-Experts from LLaMA with Continual Pre-training (EMNLP 2024)β883Updated 4 months ago
- [NeurIPS 2023] LLM-Pruner: On the Structural Pruning of Large Language Models. Support Llama-3/3.1, Llama-2, LLaMA, BLOOM, Vicuna, Baichβ¦β874Updated last month
- Best practice for training LLaMA models in Megatron-LMβ628Updated 10 months ago
- [EMNLP 2024 Industry Track] This is the official PyTorch implementation of "LLMC: Benchmarking Large Language Model Quantization with a Vβ¦β322Updated this week
- Memory optimization and training recipes to extrapolate language models' context length to 1 million tokens, with minimal hardware.β647Updated last month
- Disaggregated serving system for Large Language Models (LLMs).β359Updated 3 months ago
- Code associated with the paper **Draft & Verify: Lossless Large Language Model Acceleration via Self-Speculative Decoding**β138Updated 5 months ago
- Awesome LLM compression research papers and tools.β1,202Updated this week
- [NeurIPS 2024] KVQuant: Towards 10 Million Context Length LLM Inference with KV Cache Quantizationβ305Updated 3 months ago
- [TMLR 2024] Efficient Large Language Models: A Surveyβ1,025Updated last week
- A simple and effective LLM pruning approach.β669Updated 3 months ago
- Codes for the paper "βBench: Extending Long Context Evaluation Beyond 100K Tokens": https://arxiv.org/abs/2402.13718β285Updated last month
- Scalable toolkit for efficient model alignmentβ620Updated this week
- [ICML'24] Data and code for our paper "Training-Free Long-Context Scaling of Large Language Models"β358Updated last month
- veRL: Volcano Engine Reinforcement Learning for LLMβ318Updated this week