LMDeploy is a toolkit for compressing, deploying, and serving LLM, developed by the MMRazor and MMDeploy teams. It has the following core features: * **Efficient Inference** : LMDeploy delivers up to 1.8x higher request throughput than vLLM, by introducing key features like persistent batch(a.k.a. continuous batching), blocked KV cache, dynamic split&fuse, tensor parallelism, high-performance CUDA kernels and so on. * **Effective Quantization** : LMDeploy supports weight-only and k/v quantization, and the 4-bit inference performance is 2.4x higher than FP16. The quantization quality has been confirmed via OpenCompass evaluation. * **Effortless Distribution Server** : Leveraging the request distribution service, LMDeploy facilitates an easy and efficient deployment of multi-model services across multiple machines and cards. * **Interactive Inference Mode** : By caching the k/v of attention during multi-round dialogue processes, the engine remembers dialogue history, thus avoiding repetitive processing of historical sessions.

SGLang is a structured generation language designed for large language models (LLMs). It makes your interaction with LLMs faster and more controllable by co-designing the frontend language and the runtime system. The core features of SGLang include: - **A Flexible Front-End Language**: This allows for easy programming of LLM applications with multiple chained generation calls, advanced prompting techniques, control flow, multiple modalities, parallelism, and external interaction. - **A High-Performance Runtime with RadixAttention**: This feature significantly accelerates the execution of complex LLM programs by automatic KV cache reuse across multiple calls. It also supports other common techniques like continuous batching and tensor parallelism.

ST-LLM is a temporal-sensitive video large language model that incorporates joint spatial-temporal modeling, dynamic masking strategy, and global-local input module for effective video understanding. It has achieved state-of-the-art results on various video benchmarks. The repository provides code and weights for the model, along with demo scripts for easy usage. Users can train, validate, and use the model for tasks like video description, action identification, and reasoning.

ExecuTorch is an end-to-end solution for enabling on-device inference capabilities across mobile and edge devices including wearables, embedded devices and microcontrollers. It is part of the PyTorch Edge ecosystem and enables efficient deployment of PyTorch models to edge devices. Key value propositions of ExecuTorch are: * **Portability:** Compatibility with a wide variety of computing platforms, from high-end mobile phones to highly constrained embedded systems and microcontrollers. * **Productivity:** Enabling developers to use the same toolchains and SDK from PyTorch model authoring and conversion, to debugging and deployment to a wide variety of platforms. * **Performance:** Providing end users with a seamless and high-performance experience due to a lightweight runtime and utilizing full hardware capabilities such as CPUs, NPUs, and DSPs.

Univer is an isomorphic full-stack framework designed for creating and editing spreadsheets, documents, and slides across web and server. It is highly extensible, high-performance, and can be embedded into applications. Univer offers a wide range of features including formulas, conditional formatting, data validation, collaborative editing, printing, import & export, and more. It supports multiple languages and provides a distraction-free editing experience with a clean interface. Univer is suitable for data analysts, software developers, project managers, content creators, and educators.

Awesome-LLM-Ensemble is a collection of papers on LLM Ensemble, focusing on the comprehensive use of multiple large language models to benefit from their individual strengths. It provides a systematic review of recent developments in LLM Ensemble, including taxonomy, methods for ensemble before, during, and after inference, benchmarks, applications, and related surveys.

verl is a flexible and efficient RL training library for large language models (LLMs). It offers easy extension of diverse RL algorithms, seamless integration with existing LLM infra, flexible device mapping, and integration with popular Hugging Face models. The library provides state-of-the-art throughput, efficient actor model resharding, and supports various RL algorithms like PPO, GRPO, and more. It also supports model-based and function-based rewards for tasks like math and coding, vision-language models, and multi-modal RL. verl is used for tasks like training large language models, reasoning tasks, reinforcement learning with diverse algorithms, and multi-modal RL.

The Claude Code Ultimate Guide is an exhaustive documentation resource that takes users from beginner to power user in using Claude Code. It includes production-ready templates, workflow guides, a quiz, and a cheatsheet for daily use. The guide covers educational depth, methodologies, and practical examples to help users understand concepts and workflows. It also provides interactive onboarding, a repository structure overview, and learning paths for different user levels. The guide is regularly updated and offers a unique 257-question quiz for comprehensive assessment. Users can also find information on agent teams coverage, methodologies, annotated templates, resource evaluations, and learning paths for different roles like junior developer, senior developer, power user, and product manager/devops/designer.

A simple and convenient interface for using various neural network models. Users can interact with LLM using text, voice, and image input to generate images, videos, 3D objects, music, and audio. The tool supports a wide range of models for different tasks such as image generation, video generation, audio file separation, voice conversion, and more. Users can also view files from the outputs directory in a gallery, download models, change application settings, and check system sensors. The goal of the project is to create an easy-to-use application for utilizing neural network models.

big-AGI is an AI suite designed for professionals seeking function, form, simplicity, and speed. It offers best-in-class Chats, Beams, and Calls with AI personas, visualizations, coding, drawing, side-by-side chatting, and more, all wrapped in a polished UX. The tool is powered by the latest models from 12 vendors and open-source servers, providing users with advanced AI capabilities and a seamless user experience. With continuous updates and enhancements, big-AGI aims to stay ahead of the curve in the AI landscape, catering to the needs of both developers and AI enthusiasts.

Jarvis is a powerful virtual AI assistant designed to simplify daily tasks through voice command integration. It features automation, device management, and personalized interactions, transforming technology engagement. Built using Python and AI models, it serves personal and administrative needs efficiently, making processes seamless and productive.

The Awesome-LM-SSP repository is a collection of resources related to the trustworthiness of large models (LMs) across multiple dimensions, with a special focus on multi-modal LMs. It includes papers, surveys, toolkits, competitions, and leaderboards. The resources are categorized into three main dimensions: safety, security, and privacy. Within each dimension, there are several subcategories. For example, the safety dimension includes subcategories such as jailbreak, alignment, deepfake, ethics, fairness, hallucination, prompt injection, and toxicity. The security dimension includes subcategories such as adversarial examples, poisoning, and system security. The privacy dimension includes subcategories such as contamination, copyright, data reconstruction, membership inference attacks, model extraction, privacy-preserving computation, and unlearning.

Eridanus is a powerful data visualization tool designed to help users create interactive and insightful visualizations from their datasets. With a user-friendly interface and a wide range of customization options, Eridanus makes it easy for users to explore and analyze their data in a meaningful way. Whether you are a data scientist, business analyst, or student, Eridanus provides the tools you need to communicate your findings effectively and make data-driven decisions.

Feast is an open source feature store for machine learning, providing a fast path to manage infrastructure for productionizing analytic data. It allows ML platform teams to make features consistently available, avoid data leakage, and decouple ML from data infrastructure. Feast abstracts feature storage from retrieval, ensuring portability across different model training and serving scenarios.

AnyLabeling is a tool for effortless data labeling with AI support from YOLO and Segment Anything. It combines features from LabelImg and Labelme with an improved UI and auto-labeling capabilities. Users can annotate images with polygons, rectangles, circles, lines, and points, as well as perform auto-labeling using YOLOv5 and Segment Anything. The tool also supports text detection, recognition, and Key Information Extraction (KIE) labeling, with multiple language options available such as English, Vietnamese, and Chinese.

LMDeploy is a toolkit for compressing, deploying, and serving LLM, developed by the MMRazor and MMDeploy teams. It has the following core features: * **Efficient Inference** : LMDeploy delivers up to 1.8x higher request throughput than vLLM, by introducing key features like persistent batch(a.k.a. continuous batching), blocked KV cache, dynamic split&fuse, tensor parallelism, high-performance CUDA kernels and so on. * **Effective Quantization** : LMDeploy supports weight-only and k/v quantization, and the 4-bit inference performance is 2.4x higher than FP16. The quantization quality has been confirmed via OpenCompass evaluation. * **Effortless Distribution Server** : Leveraging the request distribution service, LMDeploy facilitates an easy and efficient deployment of multi-model services across multiple machines and cards. * **Interactive Inference Mode** : By caching the k/v of attention during multi-round dialogue processes, the engine remembers dialogue history, thus avoiding repetitive processing of historical sessions.

Open WebUI is an extensible, feature-rich, and user-friendly self-hosted WebUI designed to operate entirely offline. It supports various LLM runners, including Ollama and OpenAI-compatible APIs. For more information, be sure to check out our Open WebUI Documentation.

Mistral.rs is a fast LLM inference platform written in Rust. We support inference on a variety of devices, quantization, and easy-to-use application with an Open-AI API compatible HTTP server and Python bindings.

The llama-cpp-agent framework is a tool designed for easy interaction with Large Language Models (LLMs). Allowing users to chat with LLM models, execute structured function calls and get structured output (objects). It provides a simple yet robust interface and supports llama-cpp-python and OpenAI endpoints with GBNF grammar support (like the llama-cpp-python server) and the llama.cpp backend server. It works by generating a formal GGML-BNF grammar of the user defined structures and functions, which is then used by llama.cpp to generate text valid to that grammar. In contrast to most GBNF grammar generators it also supports nested objects, dictionaries, enums and lists of them.

BAML is a config file format for declaring LLM functions that you can then use in TypeScript or Python. With BAML you can Classify or Extract any structured data using Anthropic, OpenAI or local models (using Ollama) ## Resources  [Discord Community](https://discord.gg/boundaryml)  [Follow us on Twitter](https://twitter.com/boundaryml) * Discord Office Hours - Come ask us anything! We hold office hours most days (9am - 12pm PST). * Documentation - Learn BAML * Documentation - BAML Syntax Reference * Documentation - Prompt engineering tips * Boundary Studio - Observability and more #### Starter projects * BAML + NextJS 14 * BAML + FastAPI + Streaming ## Motivation Calling LLMs in your code is frustrating: * your code uses types everywhere: classes, enums, and arrays * but LLMs speak English, not types BAML makes calling LLMs easy by taking a type-first approach that lives fully in your codebase: 1. Define what your LLM output type is in a .baml file, with rich syntax to describe any field (even enum values) 2. Declare your prompt in the .baml config using those types 3. Add additional LLM config like retries or redundancy 4. Transpile the .baml files to a callable Python or TS function with a type-safe interface. (VSCode extension does this for you automatically). We were inspired by similar patterns for type safety: protobuf and OpenAPI for RPCs, Prisma and SQLAlchemy for databases. BAML guarantees type safety for LLMs and comes with tools to give you a great developer experience:  Jump to BAML code or how Flexible Parsing works without additional LLM calls. | BAML Tooling | Capabilities | | ----------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | BAML Compiler install | Transpiles BAML code to a native Python / Typescript library (you only need it for development, never for releases) Works on Mac, Windows, Linux  | | VSCode Extension install | Syntax highlighting for BAML files Real-time prompt preview Testing UI | | Boundary Studio open (not open source) | Type-safe observability Labeling |

WenXin-Starter is a spring-boot-starter for Baidu's "Wenxin Qianfan WENXINWORKSHOP" large model, which can help you quickly access Baidu's AI capabilities. It fully integrates the official API documentation of Wenxin Qianfan. Supports text-to-image generation, built-in dialogue memory, and supports streaming return of dialogue. Supports QPS control of a single model and supports queuing mechanism. Plugins will be added soon.

Intel® Extension for Transformers is an innovative toolkit designed to accelerate GenAI/LLM everywhere with the optimal performance of Transformer-based models on various Intel platforms, including Intel Gaudi2, Intel CPU, and Intel GPU. The toolkit provides the below key features and examples: * Seamless user experience of model compressions on Transformer-based models by extending [Hugging Face transformers](https://github.com/huggingface/transformers) APIs and leveraging [Intel® Neural Compressor](https://github.com/intel/neural-compressor) * Advanced software optimizations and unique compression-aware runtime (released with NeurIPS 2022's paper [Fast Distilbert on CPUs](https://arxiv.org/abs/2211.07715) and [QuaLA-MiniLM: a Quantized Length Adaptive MiniLM](https://arxiv.org/abs/2210.17114), and NeurIPS 2021's paper [Prune Once for All: Sparse Pre-Trained Language Models](https://arxiv.org/abs/2111.05754)) * Optimized Transformer-based model packages such as [Stable Diffusion](examples/huggingface/pytorch/text-to-image/deployment/stable_diffusion), [GPT-J-6B](examples/huggingface/pytorch/text-generation/deployment), [GPT-NEOX](examples/huggingface/pytorch/language-modeling/quantization#2-validated-model-list), [BLOOM-176B](examples/huggingface/pytorch/language-modeling/inference#BLOOM-176B), [T5](examples/huggingface/pytorch/summarization/quantization#2-validated-model-list), [Flan-T5](examples/huggingface/pytorch/summarization/quantization#2-validated-model-list), and end-to-end workflows such as [SetFit-based text classification](docs/tutorials/pytorch/text-classification/SetFit_model_compression_AGNews.ipynb) and [document level sentiment analysis (DLSA)](workflows/dlsa) * [NeuralChat](intel_extension_for_transformers/neural_chat), a customizable chatbot framework to create your own chatbot within minutes by leveraging a rich set of [plugins](https://github.com/intel/intel-extension-for-transformers/blob/main/intel_extension_for_transformers/neural_chat/docs/advanced_features.md) such as [Knowledge Retrieval](./intel_extension_for_transformers/neural_chat/pipeline/plugins/retrieval/README.md), [Speech Interaction](./intel_extension_for_transformers/neural_chat/pipeline/plugins/audio/README.md), [Query Caching](./intel_extension_for_transformers/neural_chat/pipeline/plugins/caching/README.md), and [Security Guardrail](./intel_extension_for_transformers/neural_chat/pipeline/plugins/security/README.md). This framework supports Intel Gaudi2/CPU/GPU. * [Inference](https://github.com/intel/neural-speed/tree/main) of Large Language Model (LLM) in pure C/C++ with weight-only quantization kernels for Intel CPU and Intel GPU (TBD), supporting [GPT-NEOX](https://github.com/intel/neural-speed/tree/main/neural_speed/models/gptneox), [LLAMA](https://github.com/intel/neural-speed/tree/main/neural_speed/models/llama), [MPT](https://github.com/intel/neural-speed/tree/main/neural_speed/models/mpt), [FALCON](https://github.com/intel/neural-speed/tree/main/neural_speed/models/falcon), [BLOOM-7B](https://github.com/intel/neural-speed/tree/main/neural_speed/models/bloom), [OPT](https://github.com/intel/neural-speed/tree/main/neural_speed/models/opt), [ChatGLM2-6B](https://github.com/intel/neural-speed/tree/main/neural_speed/models/chatglm), [GPT-J-6B](https://github.com/intel/neural-speed/tree/main/neural_speed/models/gptj), and [Dolly-v2-3B](https://github.com/intel/neural-speed/tree/main/neural_speed/models/gptneox). Support AMX, VNNI, AVX512F and AVX2 instruction set. We've boosted the performance of Intel CPUs, with a particular focus on the 4th generation Intel Xeon Scalable processor, codenamed [Sapphire Rapids](https://www.intel.com/content/www/us/en/products/docs/processors/xeon-accelerated/4th-gen-xeon-scalable-processors.html).

The Qianfan SDK provides best practices for large model toolchains, allowing AI workflows and AI-native applications to access the Qianfan large model platform elegantly and conveniently. The core capabilities of the SDK include three parts: large model reasoning, large model training, and general and extension: * `Large model reasoning`: Implements interface encapsulation for reasoning of Yuyan (ERNIE-Bot) series, open source large models, etc., supporting dialogue, completion, Embedding, etc. * `Large model training`: Based on platform capabilities, it supports end-to-end large model training process, including training data, fine-tuning/pre-training, and model services. * `General and extension`: General capabilities include common AI development tools such as Prompt/Debug/Client. The extension capability is based on the characteristics of Qianfan to adapt to common middleware frameworks.

h2oGPT is an Apache V2 open-source project that allows users to query and summarize documents or chat with local private GPT LLMs. It features a private offline database of any documents (PDFs, Excel, Word, Images, Video Frames, Youtube, Audio, Code, Text, MarkDown, etc.), a persistent database (Chroma, Weaviate, or in-memory FAISS) using accurate embeddings (instructor-large, all-MiniLM-L6-v2, etc.), and efficient use of context using instruct-tuned LLMs (no need for LangChain's few-shot approach). h2oGPT also offers parallel summarization and extraction, reaching an output of 80 tokens per second with the 13B LLaMa2 model, HYDE (Hypothetical Document Embeddings) for enhanced retrieval based upon LLM responses, a variety of models supported (LLaMa2, Mistral, Falcon, Vicuna, WizardLM. With AutoGPTQ, 4-bit/8-bit, LORA, etc.), GPU support from HF and LLaMa.cpp GGML models, and CPU support using HF, LLaMa.cpp, and GPT4ALL models. Additionally, h2oGPT provides Attention Sinks for arbitrarily long generation (LLaMa-2, Mistral, MPT, Pythia, Falcon, etc.), a UI or CLI with streaming of all models, the ability to upload and view documents through the UI (control multiple collaborative or personal collections), Vision Models LLaVa, Claude-3, Gemini-Pro-Vision, GPT-4-Vision, Image Generation Stable Diffusion (sdxl-turbo, sdxl) and PlaygroundAI (playv2), Voice STT using Whisper with streaming audio conversion, Voice TTS using MIT-Licensed Microsoft Speech T5 with multiple voices and Streaming audio conversion, Voice TTS using MPL2-Licensed TTS including Voice Cloning and Streaming audio conversion, AI Assistant Voice Control Mode for hands-free control of h2oGPT chat, Bake-off UI mode against many models at the same time, Easy Download of model artifacts and control over models like LLaMa.cpp through the UI, Authentication in the UI by user/password via Native or Google OAuth, State Preservation in the UI by user/password, Linux, Docker, macOS, and Windows support, Easy Windows Installer for Windows 10 64-bit (CPU/CUDA), Easy macOS Installer for macOS (CPU/M1/M2), Inference Servers support (oLLaMa, HF TGI server, vLLM, Gradio, ExLLaMa, Replicate, OpenAI, Azure OpenAI, Anthropic), OpenAI-compliant, Server Proxy API (h2oGPT acts as drop-in-replacement to OpenAI server), Python client API (to talk to Gradio server), JSON Mode with any model via code block extraction. Also supports MistralAI JSON mode, Claude-3 via function calling with strict Schema, OpenAI via JSON mode, and vLLM via guided_json with strict Schema, Web-Search integration with Chat and Document Q/A, Agents for Search, Document Q/A, Python Code, CSV frames (Experimental, best with OpenAI currently), Evaluate performance using reward models, and Quality maintained with over 1000 unit and integration tests taking over 4 GPU-hours.

Mistral.rs is a fast LLM inference platform written in Rust. We support inference on a variety of devices, quantization, and easy-to-use application with an Open-AI API compatible HTTP server and Python bindings.

Ollama is a lightweight, extensible framework for building and running language models on the local machine. It provides a simple API for creating, running, and managing models, as well as a library of pre-built models that can be easily used in a variety of applications. Ollama is designed to be easy to use and accessible to developers of all levels. It is open source and available for free on GitHub.

The llama-cpp-agent framework is a tool designed for easy interaction with Large Language Models (LLMs). Allowing users to chat with LLM models, execute structured function calls and get structured output (objects). It provides a simple yet robust interface and supports llama-cpp-python and OpenAI endpoints with GBNF grammar support (like the llama-cpp-python server) and the llama.cpp backend server. It works by generating a formal GGML-BNF grammar of the user defined structures and functions, which is then used by llama.cpp to generate text valid to that grammar. In contrast to most GBNF grammar generators it also supports nested objects, dictionaries, enums and lists of them.

This repository provides a set of ROS 2 packages to integrate llama.cpp into ROS 2. By using the llama_ros packages, you can easily incorporate the powerful optimization capabilities of llama.cpp into your ROS 2 projects by running GGUF-based LLMs and VLMs.

OneReality is a virtual waifu/assistant that you can speak to through your mic and it'll speak back to you! It has many features such as: * You can speak to her with a mic * It can speak back to you * Has short-term memory and long-term memory * Can open apps * Smarter than you * Fluent in English, Japanese, Korean, and Chinese * Can control your smart home like Alexa if you set up Tuya (more info in Prerequisites) It is built with Python, Llama-cpp-python, Whisper, SpeechRecognition, PocketSphinx, VITS-fast-fine-tuning, VITS-simple-api, HyperDB, Sentence Transformers, and Tuya Cloud IoT.

WenXin-Starter is a spring-boot-starter for Baidu's "Wenxin Qianfan WENXINWORKSHOP" large model, which can help you quickly access Baidu's AI capabilities. It fully integrates the official API documentation of Wenxin Qianfan. Supports text-to-image generation, built-in dialogue memory, and supports streaming return of dialogue. Supports QPS control of a single model and supports queuing mechanism. Plugins will be added soon.

FlexFlow Serve is an open-source compiler and distributed system for **low latency**, **high performance** LLM serving. FlexFlow Serve outperforms existing systems by 1.3-2.0x for single-node, multi-GPU inference and by 1.4-2.4x for multi-node, multi-GPU inference.