ModelTC / LightLLMView external linksLinks
LightLLM is a Python-based LLM (Large Language Model) inference and serving framework, notable for its lightweight design, easy scalability, and high-speed performance.
☆3,896Updated this week
Alternatives and similar repositories for LightLLM
Users that are interested in LightLLM are comparing it to the libraries listed below
Sorting:
- FlashInfer: Kernel Library for LLM Serving☆4,935Feb 10, 2026Updated last week
- LMDeploy is a toolkit for compressing, deploying, and serving LLMs.☆7,606Updated this week
- SGLang is a high-performance serving framework for large language models and multimodal models.☆23,547Updated this week
- Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI.☆4,771Updated this week
- TensorRT LLM provides users with an easy-to-use Python API to define Large Language Models (LLMs) and supports state-of-the-art optimizat…☆12,867Updated this week
- 📚A curated list of Awesome LLM/VLM Inference Papers with Codes: Flash-Attention, Paged-Attention, WINT8/4, Parallelism, etc.🎉☆4,990Jan 18, 2026Updated 3 weeks ago
- Transformer related optimization, including BERT, GPT☆6,392Mar 27, 2024Updated last year
- Fast and memory-efficient exact attention☆22,231Updated this week
- A throughput-oriented high-performance serving framework for LLMs☆945Oct 29, 2025Updated 3 months ago
- Medusa: Simple Framework for Accelerating LLM Generation with Multiple Decoding Heads☆2,705Jun 25, 2024Updated last year
- S-LoRA: Serving Thousands of Concurrent LoRA Adapters☆1,897Jan 21, 2024Updated 2 years ago
- MII makes low-latency and high-throughput inference possible, powered by DeepSpeed.☆2,093Jun 30, 2025Updated 7 months ago
- A high-throughput and memory-efficient inference and serving engine for LLMs☆70,205Updated this week
- [MLSys 2024 Best Paper Award] AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration☆3,436Jul 17, 2025Updated 6 months ago
- Large Language Model Text Generation Inference☆10,769Jan 8, 2026Updated last month
- FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.☆1,018Sep 4, 2024Updated last year
- Serving multiple LoRA finetuned LLM as one☆1,139May 8, 2024Updated last year
- Development repository for the Triton language and compiler☆18,429Updated this week
- Ongoing research training transformer models at scale☆15,213Updated this week
- Automatically Discovering Fast Parallelization Strategies for Distributed Deep Neural Network Training☆1,860Updated this week
- A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit and 4-bit floating point (FP8 and FP4) precision on H…☆3,163Updated this week
- [MLSys'25] QServe: W4A8KV4 Quantization and System Co-design for Efficient LLM Serving; [MLSys'25] LServe: Efficient Long-sequence LLM Se…☆812Mar 6, 2025Updated 11 months ago
- RTP-LLM: Alibaba's high-performance LLM inference engine for diverse applications.☆1,049Updated this week
- An easy-to-use LLMs quantization package with user-friendly apis, based on GPTQ algorithm.☆5,028Apr 11, 2025Updated 10 months ago
- Efficient Triton Kernels for LLM Training☆6,141Updated this week
- fastllm是后端无依赖的高性能大模型推理库。同时支持张量并行推理稠密模型和混合模式推理MOE模型，任意10G以上显卡即可推理满血DeepSeek。双路9004/9005服务器+单显卡部署DeepSeek满血满精度原版模型，单并发20tps；INT4量化模型单并发30tp…☆4,149Updated this week
- Disaggregated serving system for Large Language Models (LLMs).☆776Apr 6, 2025Updated 10 months ago
- Distributed Compiler based on Triton for Parallel Systems☆1,350Feb 9, 2026Updated last week
- A fast communication-overlapping library for tensor/expert parallelism on GPUs.☆1,247Aug 28, 2025Updated 5 months ago
- [ICML 2024] Break the Sequential Dependency of LLM Inference Using Lookahead Decoding☆1,316Mar 6, 2025Updated 11 months ago
- Official Implementation of EAGLE-1 (ICML'24), EAGLE-2 (EMNLP'24), and EAGLE-3 (NeurIPS'25).☆2,182Jan 27, 2026Updated 2 weeks ago
- [ICLR 2024] Efficient Streaming Language Models with Attention Sinks☆7,188Jul 11, 2024Updated last year
- Accessible large language models via k-bit quantization for PyTorch.☆7,952Updated this week
- verl: Volcano Engine Reinforcement Learning for LLMs☆19,132Updated this week
- how to optimize some algorithm in cuda.☆2,819Updated this week
- AutoAWQ implements the AWQ algorithm for 4-bit quantization with a 2x speedup during inference. Documentation:☆2,314May 11, 2025Updated 9 months ago
- 🚀 Efficient implementations of state-of-the-art linear attention models☆4,379Updated this week
- Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels☆5,191Updated this week
- [NeurIPS'24 Spotlight, ICLR'25, ICML'25] To speed up Long-context LLMs' inference, approximate and dynamic sparse calculate the attention…☆1,188Sep 30, 2025Updated 4 months ago