AI & ML interests

None defined yet.

EmbeddedLLM

About EmbeddedLLM

EmbeddedLLM is an open-source company dedicated to advancing the field of Large Language Models (LLMs) through innovative backend solutions and hardware optimizations. Our mission is to make powerful generative models work on all platforms, from edge to private cloud, ensuring accessibility and efficiency for a wide range of applications.

Highlighted Repositories

  1. EmbeddedLLM/JamAIBase
  • Description: JamAI Base is an open-source RAG (Retrieval-Augmented Generation) backend platform that integrates an embedded database (SQLite) and an embedded vector database (LanceDB) with managed memory and RAG capabilities. It features built-in LLM, vector embeddings, and reranker orchestration and management, all accessible through a convenient, intuitive, spreadsheet-like UI and a simple REST API.
  • Key Features:
    • Embedded database (SQLite) and vector database (LanceDB)
    • Managed memory and RAG capabilities
    • Built-in LLM, vector embeddings, and reranker orchestration
    • Intuitive spreadsheet-like UI
    • Simple REST API
  1. EmbeddedLLM/vllm-rocm
  • Description: This repository is a port of vLLM for AMD GPUs, providing a high-throughput and memory-efficient inference and serving engine for LLMs optimized for ROCm.
  • Key Features:
    • Vision Language Models support
    • New features not yet available in the upstream
    • Optimized for AMD GPUs with ROCm support
  1. EmbeddedLLM/embeddedllm
  • Description: It is a AIPC embedded LLM Engine unifying and provide stable way to run LLM fast on CPU, iGPU, GPU. It supports launching OpenAI-API-Compatible API server powered by our engine.
  • Key Features:
    • Supported hardwares: CPU (ONNX), AMD iGPU (ONNX-DirectML), Intel iGPU (IPEX-LLM, OpenVINO), Intel XPU (IPEX-LLM, OpenVINO), Nvidia GPU (ONNX-CUDA).
    • Provide prebuilt, ready-to-run Windows 11 executable.
    • Vision Language Models support (CPU)

Join Us

We invite you to explore our repositories and models, contribute to our projects, and join us in pushing the boundaries of what's possible with LLMs.

datasets

None public yet