A unified library of SOTA model optimization techniques like quantization, distillation, pruning, neural architecture search, speculative decoding, etc. It compresses deep learning models for downstream deployment frameworks like TensorRT-LLM, TensorRT, vLLM, etc. to optimize inference speed.
☆5,243Oct 8, 2026Updated this week
Alternatives and similar repositories for Model-Optimizer
Users that are interested in Model-Optimizer are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.
Sorting:
- State-of-the-art LLM compression, built for production inference with vLLM☆3,859Updated this week
- FlashInfer: Kernel Library for LLM Serving☆6,556Updated 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…☆3,571Updated 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…☆14,777Updated this week
- PyTorch native quantization for training and inference☆2,995Updated this week
- Serverless GPU API endpoints on Runpod - Get Bonus Credits • AdSkip the infrastructure headaches. Auto-scaling, pay-as-you-go, no-ops approach lets you focus on innovating your application.
- A Datacenter Scale Distributed Inference Serving Framework☆8,247Updated this week
- Train speculative decoding models effortlessly and port them smoothly to SGLang serving.☆1,204Updated this week
- CUDA Templates and Python DSLs for High-Performance Linear Algebra☆10,546Sep 23, 2026Updated 2 weeks ago
- SGLang is a high-performance serving framework for large language models and multimodal models.☆36,875Updated this week
- Model Compression Toolbox for Large Language Models and Diffusion Models☆801Aug 14, 2025Updated last year
- High-performance, light-weight C++ LLM and VLM Inference Software for Physical AI☆583Updated this week
- NVIDIA® TensorRT™ is an SDK for high-performance deep learning inference on NVIDIA GPUs. This repository contains the open source compone…☆13,391Sep 22, 2026Updated 2 weeks ago
- FP16xINT4 LLM inference kernel that can achieve near-ideal ~4x speedups up to medium batchsizes of 16-32 tokens.☆1,153Sep 4, 2024Updated 2 years ago
- Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels☆8,516Updated this week
- GPUs on demand by Runpod - Special Offer Available • AdRun AI, ML, and HPC workloads on powerful cloud GPUs—without limits or wasted spend. Deploy GPUs in under a minute and pay by the second.
- DeepGEMM: clean and efficient BLAS kernel library on GPU☆8,874Sep 30, 2026Updated last week
- Tile primitives for speedy kernels☆3,746Sep 12, 2026Updated 3 weeks ago
- Official Implementation of EAGLE-1 (ICML'24), EAGLE-2 (EMNLP'24), and EAGLE-3 (NeurIPS'25).☆2,548Feb 20, 2026Updated 7 months ago
- A high-throughput and memory-efficient inference and serving engine for LLMs☆93,397Updated this week
- Transformer related optimization, including BERT, GPT☆6,454Mar 27, 2024Updated 2 years ago
- Fast and memory-efficient exact attention☆25,106Updated this week
- [ICLR2025, ICML2025, NeurIPS2025 Spotlight] Quantized Attention achieves speedup of 2-5x compared to FlashAttention, without losing end-t…☆3,963Jan 17, 2026Updated 8 months ago
- [MLSys 2024 Best Paper Award] AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration☆3,649Jul 17, 2025Updated last year
- Mirage Persistent Kernel: Compiling LLMs into a MegaKernel☆2,542Updated this week
- GPU virtual machines on DigitalOcean Gradient AI • AdGet to production fast with high-performance AMD and NVIDIA GPUs you can spin up in seconds. The definition of operational simplicity.
- LLM model quantization (compression) toolkit with HW acceleration support for Nvidia, AMD, Intel GPU and Intel/AMD/Apple CPU via HF, vLLM…☆1,270Updated this week
- Development repository for the Triton language and compiler☆20,321Updated this week
- The Triton TensorRT-LLM Backend☆944Updated this week
- Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI.☆6,737Updated this week
- 🚀 Efficient implementations for emerging model architectures☆5,833Updated this week
- A simple and effective post training quantization toolkit for high-accuracy low-bit LLM inference|简洁且高效的后训练量化工具包☆1,629Updated this week
- PyTorch/TorchScript/FX compiler for NVIDIA GPUs using TensorRT☆3,009Updated this week
- An Open-source book with Modern CUDA learning Notes for Beginners - BF16/FP8/FP4, HGEMM, CuTe, Flash-Attention, etc.☆12,047Updated this week
- A safetensors extension to efficiently store sparse quantized tensors on disk☆329Updated this week
- 1-Click AI Models by DigitalOcean Gradient • AdDeploy popular AI models on DigitalOcean Gradient GPU virtual machines with just a single click. Zero configuration with optimized deployments.
- Fast low-bit matmul kernels in Triton☆488Oct 1, 2026Updated last week
- [MLSys'25] QServe: W4A8KV4 Quantization and System Co-design for Efficient LLM Serving; [MLSys'25] LServe: Efficient Long-sequence LLM Se…☆862Mar 6, 2025Updated last year
- Scalable toolkit for efficient model reinforcement☆2,052Updated this week
- PPL Quantization Tool (PPQ) is a powerful offline neural network quantization tool.☆1,822Mar 28, 2024Updated 2 years ago
- LMCache: Supercharge Your LLM with the Fastest KV Cache Layer☆11,979Updated this week
- Distributed Compiler and Optimized Parallel Kernels☆1,559Sep 18, 2026Updated 3 weeks ago
- [ICML 2023] SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models☆1,695Jul 12, 2024Updated 2 years ago