microsoft / MInferenceLinks
[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.
☆1,152Updated last month
Alternatives and similar repositories for MInference
Users that are interested in MInference are comparing it to the libraries listed below
Sorting:
- A throughput-oriented high-performance serving framework for LLMs☆915Updated 3 weeks ago
- [ICML 2024] Break the Sequential Dependency of LLM Inference Using Lookahead Decoding☆1,305Updated 8 months ago
- [MLSys'25] QServe: W4A8KV4 Quantization and System Co-design for Efficient LLM Serving; [MLSys'25] LServe: Efficient Long-sequence LLM Se…☆781Updated 8 months ago
- Memory optimization and training recipes to extrapolate language models' context length to 1 million tokens, with minimal hardware.☆750Updated last year
- FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.☆954Updated last year
- [ICLR 2025] DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads☆505Updated 9 months ago
- Serving multiple LoRA finetuned LLM as one☆1,120Updated last year
- OLMoE: Open Mixture-of-Experts Language Models☆908Updated 2 months ago
- For releasing code related to compression methods for transformers, accompanying our publications☆451Updated 10 months ago
- Fast, Flexible and Portable Structured Generation☆1,391Updated last week
- Transformers-compatible library for applying various compression algorithms to LLMs for optimized deployment with vLLM☆2,266Updated this week
- [ICLR2024 spotlight] OmniQuant is a simple and powerful quantization technique for LLMs.☆875Updated 6 months ago
- LLM model quantization (compression) toolkit with hw acceleration support for Nvidia CUDA, AMD ROCm, Intel XPU and Intel/AMD/Apple CPU vi…☆886Updated last week
- The code of our paper "InfLLM: Unveiling the Intrinsic Capacity of LLMs for Understanding Extremely Long Sequences with Training-Free Mem…☆388Updated last year
- Fast inference from large lauguage models via speculative decoding☆856Updated last year
- Ring attention implementation with flash attention☆921Updated 2 months ago
- VPTQ, A Flexible and Extreme low-bit quantization algorithm☆668Updated 7 months ago
- Efficient LLM Inference over Long Sequences☆391Updated 5 months ago
- Train speculative decoding models effortlessly and port them smoothly to SGLang serving.☆483Updated last week
- [ICLR 2024] Sheared LLaMA: Accelerating Language Model Pre-training via Structured Pruning☆632Updated last year
- Official Implementation of EAGLE-1 (ICML'24), EAGLE-2 (EMNLP'24), and EAGLE-3 (NeurIPS'25).☆2,003Updated 2 weeks ago
- Advanced quantization toolkit for LLMs and VLMs. Native support for WOQ, MXFP4, NVFP4, GGUF, Adaptive Bits and seamless integration with …☆724Updated this week
- LongBench v2 and LongBench (ACL 25'&24')☆1,020Updated 10 months ago
- Analyze the inference of Large Language Models (LLMs). Analyze aspects like computation, storage, transmission, and hardware roofline mod…☆581Updated last year
- Medusa: Simple Framework for Accelerating LLM Generation with Multiple Decoding Heads☆2,664Updated last year
- ⛷ ️ LLaMA-MoE: Building Mixture-of-Experts from LLaMA with Continual Pre-training (EMNLP 2024)☆994Updated 11 months ago
- A high-performance inference system for large language models, designed for production environments.☆485Updated 3 weeks ago
- [EMNLP 2024 & AAAI 2026] A powerful toolkit for compressing large models including LLM, VLM, and video generation models.☆625Updated last week
- Muon is Scalable for LLM Training☆1,365Updated 3 months ago
- LLM KV cache compression made easy☆694Updated this week