microsoft / MInference
[NeurIPS'24 Spotlight, ICLR'25, ICML'25] To speed up Long-context LLMs' inference, approximate and dynamic sparse calculate the attention, which reduces inference latency by up to 10x for pre-filling on an A100 while maintaining accuracy.
☆1,005Updated this week
Alternatives and similar repositories for MInference:
Users that are interested in MInference are comparing it to the libraries listed below
- Official Implementation of EAGLE-1 (ICML'24), EAGLE-2 (EMNLP'24), and EAGLE-3.☆1,211Updated this week
- A throughput-oriented high-performance serving framework for LLMs☆805Updated last week
- Transformers-compatible library for applying various compression algorithms to LLMs for optimized deployment with vLLM☆1,301Updated this week
- [ICML 2024] Break the Sequential Dependency of LLM Inference Using Lookahead Decoding☆1,243Updated 2 months ago
- [MLSys'25] QServe: W4A8KV4 Quantization and System Co-design for Efficient LLM Serving; [MLSys'25] LServe: Efficient Long-sequence LLM Se…☆663Updated 2 months ago
- VPTQ, A Flexible and Extreme low-bit quantization algorithm☆632Updated last week
- Memory optimization and training recipes to extrapolate language models' context length to 1 million tokens, with minimal hardware.☆720Updated 7 months ago
- FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.☆812Updated 8 months ago
- OLMoE: Open Mixture-of-Experts Language Models☆739Updated last month
- Advanced Quantization Algorithm for LLMs/VLMs.☆449Updated last week
- Fast inference from large lauguage models via speculative decoding☆722Updated 8 months ago
- [ICLR2024 spotlight] OmniQuant is a simple and powerful quantization technique for LLMs.☆805Updated 7 months ago
- The code of our paper "InfLLM: Unveiling the Intrinsic Capacity of LLMs for Understanding Extremely Long Sequences with Training-Free Mem…☆354Updated last year
- Official implementation of Half-Quadratic Quantization (HQQ)☆807Updated this week
- Fast, Flexible and Portable Structured Generation☆922Updated this week
- Serving multiple LoRA finetuned LLM as one☆1,058Updated last year
- For releasing code related to compression methods for transformers, accompanying our publications☆425Updated 3 months ago
- [ICLR 2024] Sheared LLaMA: Accelerating Language Model Pre-training via Structured Pruning☆607Updated last year
- FlashInfer: Kernel Library for LLM Serving☆2,788Updated this week
- [ICLR 2025] Alignment Data Synthesis from Scratch by Prompting Aligned LLMs with Nothing. Your efficient and high-quality synthetic data …☆693Updated last month
- [ICLR 2025] DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads☆456Updated 2 months ago
- Medusa: Simple Framework for Accelerating LLM Generation with Multiple Decoding Heads☆2,516Updated 10 months ago
- Analyze the inference of Large Language Models (LLMs). Analyze aspects like computation, storage, transmission, and hardware roofline mod…☆450Updated 7 months ago
- 📰 Must-read papers and blogs on Speculative Decoding ⚡️☆714Updated this week
- A family of compressed models obtained via pruning and knowledge distillation☆336Updated 5 months ago
- ⛷️ LLaMA-MoE: Building Mixture-of-Experts from LLaMA with Continual Pre-training (EMNLP 2024)☆960Updated 5 months ago
- ☆924Updated 3 months ago
- AutoAWQ implements the AWQ algorithm for 4-bit quantization with a 2x speedup during inference. Documentation:☆2,145Updated this week
- Muon is Scalable for LLM Training☆1,039Updated last month
- Large Reasoning Models☆804Updated 5 months ago