Open-source alternatives to FlexGen

FlexLLMGen is an inference engine and runtime designed to run large language models on a single GPU by combining weight compression with tensor offloading. It reduces model weight memory usage by approximately 70% through 4-bit quantization, and stores model parameters, attention cache, and hidden states across GPU, CPU, and disk to fit models larger than available GPU memory. The project distinguishes itself through a throughput-oriented batching approach that processes multiple generation requests together in large batches to maximize throughput on a single GPU. It also supports distributed

MiniCPM is a collection of small language models designed for local, on-device deployment in resource-constrained environments. The project focuses on running dense Transformer models on consumer hardware, including GPUs, CPUs, and Apple Silicon, without requiring custom code forks. The project distinguishes itself through heavy optimization for edge hardware, utilizing quantized weight compression in GGUF and MLX formats to reduce memory overhead. It implements advanced inference techniques such as speculative sampling and radix-tree prefix caching to accelerate generation speed and throughp

llm-compressor is a quantization toolkit and post-training library designed to reduce the memory footprint and size of large language models. It provides a framework for compressing models using weight and activation quantization to enable more efficient deployment. The project distinguishes itself through a distributed quantization framework that utilizes data-parallel processing and disk-based weight offloading to handle massive model checkpoints that exceed available system memory. It includes specialized compressors for diverse architectures, including Mixture-of-Experts, Vision-Language,

KoboldCPP is a local large language model inference engine and GGUF model runner designed to execute quantized models on personal hardware. It functions as a multimodal AI server and API gateway, providing OpenAI-compatible endpoints that allow third-party clients to interact with locally hosted models. The project distinguishes itself as an AI storytelling backend, featuring dedicated tools for long-form narrative management through persistent memory, world lore tracking, and character state management. It further extends its capabilities as a multimodal server capable of processing text, im

Qwen-Image is a text-to-image model and large language model image generation framework. It functions as an AI image editing suite and a personalized image trainer, capable of producing high-fidelity visuals and accurate typography from natural language descriptions. The system is distinguished by its precision text rendering engine, which integrates multi-script calligraphy and layout-coherent alphabetic text into images. It provides specialized capabilities for subject identity preservation and consistent subject generation across different poses and viewpoints, alongside a training pipelin

lmdeploy is a high-performance inference engine and deployment framework for large language models and vision models. It functions as a multi-modal model server and compression toolkit designed to serve models with high throughput and low latency. The system enables the distribution of model services across multiple machines using request-based load balancing and tensor parallelism. It includes specialized tools for model quantization and compression to reduce the memory footprint of weights and caches. The framework covers broad capability areas including production deployment, distributed

OpenVINO is an AI inference engine and model serving platform designed to execute optimized deep learning models across CPUs, GPUs, and NPUs through a unified API. It includes a model optimization toolkit for converting, quantizing, and compressing models from various frameworks, alongside a specialized generative AI runtime for large language models. The project distinguishes itself through a plugin-based hardware acceleration layer that maps neural network operations to vendor-specific drivers. It features advanced execution mechanisms such as continuous batching, speculative decoding, and

TensorRT-LLM is a platform and toolkit designed for compiling, optimizing, and serving transformer-based models on accelerated hardware. It functions as a framework that transforms machine learning models into efficient execution graphs, providing an engine to refine these models for specific hardware to maximize throughput and minimize latency during text generation. The project distinguishes itself through advanced execution strategies that manage the entire inference pipeline. It utilizes kernel-level fusion and static graph execution to optimize mathematical operations and computational f

llama.cpp is a high-performance C++ inference engine and runtime for executing large language models locally across various hardware architectures. It provides the core components for local model execution, including a dedicated model quantizer for compressing weights into the GGUF format and a system for generating text embeddings for semantic search. The project distinguishes itself through specialized memory and execution optimizations, such as block-wise weight quantization to reduce memory footprints and memory-mapped model loading. It supports structured text generation by using formal

Sglang is a high-performance inference engine and serving system designed for large language and multimodal models. It provides a programmable interface for orchestrating complex generation workflows, enabling developers to coordinate multi-turn dialogues, tool invocations, and reasoning chains through a domain-specific language. The platform is built to support production-scale deployments, offering an OpenAI-compatible API that allows for integration with existing application ecosystems. The system distinguishes itself through a disaggregated architecture that separates compute-intensive pr

vLLM is a high-throughput inference engine designed for the efficient serving and execution of large language models. It functions as a production-ready distributed model server, providing standard API protocols for online serving while also supporting offline batch processing. The system is built to maximize token generation speed and memory efficiency, enabling both large-scale cloud deployments and local execution on personal hardware. The project distinguishes itself through advanced memory management and request scheduling techniques, most notably its use of non-contiguous key-value cach

Intel XPU LLM Acceleration Library is a toolkit designed to accelerate large language model inference and finetuning on Intel CPUs, GPUs, and NPUs. It provides a distributed inference engine for scaling models across multiple accelerators, a multimodal model runtime for vision and speech tasks, and a low-bit model quantization tool for converting weights into INT4, FP8, and GGUF formats. The project features a parameter-efficient finetuning framework that enables model adaptation using QLoRA and DPO on Intel hardware. It distinguishes itself by providing specialized optimizations for Intel XP

AISystem is a comprehensive AI full-stack infrastructure project covering the entire pipeline from AI chip architecture to high-level training frameworks. It encompasses the development of AI compiler frameworks, inference engines, and distributed training orchestrators designed to coordinate workloads across a heterogeneous compute stack of CPUs, GPUs, and NPUs. The project focuses on the deep integration of software and hardware, employing software-hardware co-design to align tensor layouts with physical memory structures. It provides specialized capabilities for accelerating Transformer mo

Neural Compressor is a deep learning model compression toolkit and AI inference acceleration engine. It functions as an automated model quantization tool and hardware-aware model compiler designed to reduce the memory footprint of neural networks and decrease execution latency. The project provides specialized frameworks for optimizing large language models, utilizing weight-only quantization and hardware-specific kernels to improve the operational efficiency of generative AI workloads. It maps neural network operators to specialized CPU and GPU vector instructions to accelerate model executi

ComfyUI is a modular generative AI workflow orchestrator and node-based GUI for designing and executing complex diffusion model pipelines. It functions as both a visual interface for building generative logic graphs and a programmable backend API that exposes diffusion model operations for external integration. The system distinguishes itself through a graph-based execution model that supports differential workflow execution, re-running only modified nodes to reduce computation. It features dynamic model offloading to manage memory between system RAM and GPU VRAM and utilizes metadata-embedde

Turbopilot is a local large language model inference server designed to provide private code completions. It functions as a self-hosted engine that executes models on local hardware, ensuring development workflows remain offline and source code does not leave the machine. The system includes a quantization tool and model manager used to compress weights and merge sharded data into a unified binary format. This optimization reduces memory footprints and accelerates loading for execution on consumer-grade hardware. Performance is managed through a GPU accelerated inference engine that offloads

This project is a comprehensive technical course study guide and reference for learning the architectures and training methods of Transformers and large language models. It serves as a technical overview for understanding how neural networks process data and how to align model behavior with specific performance goals. The repository provides specialized guides on several key areas of model development. This includes detailed references for transformer architectures, implementation frameworks for retrieval-augmented generation and agentic workflows, and technical guides for model optimization

This project provides a comprehensive technical guide and framework for engineering large-scale machine learning systems. It covers the full lifecycle of model development, focusing on the infrastructure and computational principles required to build, train, and serve generative AI models across distributed GPU clusters. The repository distinguishes itself by offering deep-dive tutorials and implementation strategies for complex system challenges. It emphasizes high-performance architectural primitives, such as collective communication orchestration, distributed tensor sharding, and static gr

This library provides a framework for parameter-efficient fine-tuning, enabling the adaptation of large pretrained models by training only a small subset of parameters. It functions as a distributed model training system and optimization toolkit, designed to reduce the computational and memory requirements typically associated with full model fine-tuning. The project distinguishes itself through a suite of methods for modular adapter composition, including low-rank matrix decomposition and activation-based scaling. It supports the integration of multiple task-specific adapter modules, allowin

Llama 3 is a collection of pretrained, autoregressive transformer-based models designed for natural language generation, reasoning, and complex instruction following. It functions as a generative AI framework that provides the infrastructure for managing model weights, executing neural network inference, and handling computational workloads across diverse knowledge domains. The project distinguishes itself through an integrated AI safety toolkit that employs secondary classification filtering to inspect inputs and outputs, ensuring adherence to usage compliance and safety standards. It suppor

FasterTransformer is a high-performance inference optimization library and distributed runtime designed to accelerate the execution of transformer models. It provides a toolkit for reducing model precision and parallelizing execution across multiple GPUs to increase throughput and reduce latency for large language models. The framework utilizes a C++ backend with custom CUDA kernels to replace generic operations with optimized GPU instructions. It implements tensor and pipeline parallelism to shard model weights and distribute compute operations across multiple devices. The system includes c

gpt-fast is a PyTorch transformer inference engine designed for low-latency text generation. It functions as a distributed GPU inference library, a quantized model runner, and a speculative decoding framework. The system utilizes a speculative decoding workflow where a small draft model predicts token sequences for verification by a larger model to accelerate generation. It supports quantized model execution to reduce memory footprint and implements tensor parallelism to split computations across multiple GPUs. The project includes a standardized evaluation harness to measure the accuracy an

LLamaSharp is a .NET LLM inference library and local runtime that enables the execution of large language models on CPU and GPU hardware. It serves as a multimodal AI library capable of processing both text and image inputs to generate analytical textual responses without relying on external APIs. The project distinguishes itself as a grammar-based text generator that enforces specific output formats, such as JSON, through constrained sampling pipelines. It also functions as a retrieval augmented generation framework integration, allowing the combination of local inference with external data

Text Embeddings Inference is a high-performance inference server designed to host text embedding and sequence classification models as scalable API endpoints. It provides a vector embedding API to convert text into dense representations and a cross-encoder reranking server for scoring the relevance of document sequences against a query. The project features a GPU-accelerated inference engine that utilizes dynamic batching and specialized kernels to maximize throughput. It offers a high-performance binary interface via gRPC as an alternative to standard HTTP to reduce network latency and seria

Sana is a framework for high-resolution image and video synthesis based on a linear diffusion transformer. It provides a toolkit for the training, fine-tuning, and execution of text-to-image and text-to-video models, as well as a video generative world model capable of simulating physical environments with precise spatial control. The project is distinguished by its use of linear complexity layers to handle high resolutions and its support for long-form, minute-length video generation in real time. It implements a two-stage inference paradigm that separates structural generation from visual t

Personaplex is an LLM speech-to-speech framework and conversational AI persona engine designed for real-time voice interfaces. It provides a system for defining AI identities and vocal characteristics through a combination of text-based role prompts and audio reference files. The project features a real-time AI voice interface that supports full-duplex human-AI dialogue, enabling multiple parties to speak and listen simultaneously via bidirectional audio streaming. It includes a GPU-accelerated audio processor and a speech-to-speech pipeline to facilitate low-latency conversations. The frame